feat: update UI for showing a diff

Option+Left or Option+Right should move cursor to beginning/end of the
word.

We weren't listening to what terminals are sending (on MacOS) and were
therefore printing b or f instead of moving cursor. We were actually in
the first match clause and returning char insertion
(https://github.com/openai/codex/pull/1895/files#diff-6bf130cd00438cc27a38c5a4d9937a27cf9a324c191de4b74fc96019d362be6dL209)

Tested on Apple Terminal, iTerm, Ghostty

.github/actions/codex/src/process-label.ts

@@ -91,7 +91,38 @@ async function processLabel(
   labelConfig: LabelConfig,
 ): Promise<void> {
   const template = labelConfig.getPromptTemplate();
-  const populatedTemplate = await renderPromptTemplate(template, ctx);
+
+  // If this is a review label, prepend explicit PR-diff scoping guidance to
+  // reduce out-of-scope feedback. Do this before rendering so placeholders in
+  // the guidance (e.g., {CODEX_ACTION_GITHUB_EVENT_PATH}) are substituted.
+  const isReview = label.toLowerCase().includes("review");
+  const reviewScopeGuidance = `
+PR Diff Scope
+- Only review changes between the PR's merge-base and head; do not comment on commits or files outside this range.
+- Derive the base/head SHAs from the event JSON at {CODEX_ACTION_GITHUB_EVENT_PATH}, then compute and use the PR diff for all analysis and comments.
+
+Commands to determine scope
+- Resolve SHAs:
+  - BASE_SHA=$(jq -r '.pull_request.base.sha // .pull_request.base.ref' "{CODEX_ACTION_GITHUB_EVENT_PATH}")
+  - HEAD_SHA=$(jq -r '.pull_request.head.sha // .pull_request.head.ref' "{CODEX_ACTION_GITHUB_EVENT_PATH}")
+  - BASE_SHA=$(git rev-parse "$BASE_SHA")
+  - HEAD_SHA=$(git rev-parse "$HEAD_SHA")
+- Prefer triple-dot (merge-base) semantics for PR diffs:
+  - Changed commits: git log --oneline "$BASE_SHA...$HEAD_SHA"
+  - Changed files: git diff --name-status "$BASE_SHA...$HEAD_SHA"
+  - Review hunks: git diff -U0 "$BASE_SHA...$HEAD_SHA"
+
+Review rules
+- Anchor every comment to a file and hunk present in git diff "$BASE_SHA...$HEAD_SHA".
+- If you mention context outside the diff, label it as "Follow-up (outside this PR scope)" and keep it brief (<=2 bullets).
+- Do not critique commits or files not reachable in the PR range (merge-base(base, head) → head).
+`.trim();
+
+  const effectiveTemplate = isReview
+    ? `${reviewScopeGuidance}\n\n${template}`
+    : template;
+
+  const populatedTemplate = await renderPromptTemplate(effectiveTemplate, ctx);
 
   // Always run Codex and post the resulting message as a comment.
   let commentBody = await runCodex(populatedTemplate, ctx);

.github/codex/labels/codex-rust-review.md

@@ -6,18 +6,134 @@ Then provide the **review** (1-2 sentences plus bullet points, friendly tone).
 
 Things to look out for when doing the review:
 
+## General Principles
+
 - **Make sure the pull request body explains the motivation behind the change.** If the author has failed to do this, call it out, and if you think you can deduce the motivation behind the change, propose copy.
 - Ideally, the PR body also contains a small summary of the change. For small changes, the PR title may be sufficient.
 - Each PR should ideally do one conceptual thing. For example, if a PR does a refactoring as well as introducing a new feature, push back and suggest the refactoring be done in a separate PR. This makes things easier for the reviewer, as refactoring changes can often be far-reaching, yet quick to review.
-- If the nature of the change seems to have a visual component (which is often the case for changes to `codex-rs/tui`), recommend including a screenshot or video to demonstrate the change, if appropriate.
-- Rust files should generally be organized such that the public parts of the API appear near the top of the file and helper functions go below. This is analagous to the "inverted pyramid" structure that is favored in journalism.
-- Encourage the use of small enums or the newtype pattern in Rust if it helps readability without adding significant cognitive load or lines of code.
-- Be wary of large files and offer suggestions for how to break things into more reasonably-sized files.
-- When modifying a `Cargo.toml` file, make sure that dependency lists stay alphabetically sorted. Also consider whether a new dependency is added to the appropriate place (e.g., `[dependencies]` versus `[dev-dependencies]`)
-- If you see opportunities for the changes in a diff to use more idiomatic Rust, please make specific recommendations. For example, favor the use of expressions over `return`.
 - When introducing new code, be on the lookout for code that duplicates existing code. When found, propose a way to refactor the existing code such that it should be reused.
+
+## Code Organization
+
 - Each create in the Cargo workspace in `codex-rs` has a specific purpose: make a note if you believe new code is not introduced in the correct crate.
 - When possible, try to keep the `core` crate as small as possible. Non-core but shared logic is often a good candidate for `codex-rs/common`.
+- Be wary of large files and offer suggestions for how to break things into more reasonably-sized files.
+- Rust files should generally be organized such that the public parts of the API appear near the top of the file and helper functions go below. This is analagous to the "inverted pyramid" structure that is favored in journalism.
+
+## Assertions in Tests
+
+Assert the equality of the entire objects instead of doing "piecemeal comparisons," performing `assert_eq!()` on individual fields.
+
+Note that unit tests also function as "executable documentation." As shown in the following example, "piecemeal comparisons" are often more verbose, provide less coverage, and are not as useful as executable documentation.
+
+For example, suppose you have the following enum:
+
+```rust
+#[derive(Debug, PartialEq)]
+enum Message {
+    Request {
+        id: String,
+        method: String,
+        params: Option<serde_json::Value>,
+    },
+    Notification {
+        method: String,
+        params: Option<serde_json::Value>,
+    },
+}
+```
+
+This is an example of a _piecemeal_ comparison:
+
+```rust
+// BAD: Piecemeal Comparison
+
+#[test]
+fn test_get_latest_messages() {
+    let messages = get_latest_messages();
+    assert_eq!(messages.len(), 2);
+
+    let m0 = &messages[0];
+    match m0 {
+        Message::Request { id, method, params } => {
+            assert_eq!(id, "123");
+            assert_eq!(method, "subscribe");
+            assert_eq!(
+                *params,
+                Some(json!({
+                    "conversation_id": "x42z86"
+                }))
+            )
+        }
+        Message::Notification { .. } => {
+            panic!("expected Request");
+        }
+    }
+
+    let m1 = &messages[1];
+    match m1 {
+        Message::Request { .. } => {
+            panic!("expected Notification");
+        }
+        Message::Notification { method, params } => {
+            assert_eq!(method, "log");
+            assert_eq!(
+                *params,
+                Some(json!({
+                    "level": "info",
+                    "message": "subscribed"
+                }))
+            )
+        }
+    }
+}
+```
+
+This is a _deep_ comparison:
+
+```rust
+// GOOD: Verify the entire structure with a single assert_eq!().
+
+use pretty_assertions::assert_eq;
+
+#[test]
+fn test_get_latest_messages() {
+    let messages = get_latest_messages();
+
+    assert_eq!(
+        vec![
+            Message::Request {
+                id: "123".to_string(),
+                method: "subscribe".to_string(),
+                params: Some(json!({
+                    "conversation_id": "x42z86"
+                })),
+            },
+            Message::Notification {
+                method: "log".to_string(),
+                params: Some(json!({
+                    "level": "info",
+                    "message": "subscribed"
+                })),
+            },
+        ],
+        messages,
+    );
+}
+```
+
+## More Tactical Rust Things To Look Out For
+
+- Do not use `unsafe` (unless you have a really, really good reason like using an operating system API directly and no safe wrapper exists). For example, there are cases where it is tempting to use `unsafe` in order to use `std::env::set_var()`, but this indeed `unsafe` and has led to race conditions on multiple occasions. (When this happens, find a mechanism other than environment variables to use for configuration.)
+- Encourage the use of small enums or the newtype pattern in Rust if it helps readability without adding significant cognitive load or lines of code.
+- If you see opportunities for the changes in a diff to use more idiomatic Rust, please make specific recommendations. For example, favor the use of expressions over `return`.
+- When modifying a `Cargo.toml` file, make sure that dependency lists stay alphabetically sorted. Also consider whether a new dependency is added to the appropriate place (e.g., `[dependencies]` versus `[dev-dependencies]`)
+
+## Pull Request Body
+
+- If the nature of the change seems to have a visual component (which is often the case for changes to `codex-rs/tui`), recommend including a screenshot or video to demonstrate the change, if appropriate.
 - References to existing GitHub issues and PRs are encouraged, where appropriate, though you likely do not have network access, so may not be able to help here.
 
+# PR Information
+
 {CODEX_ACTION_GITHUB_EVENT_PATH} contains the JSON that triggered this GitHub workflow. It contains the `base` and `head` refs that define this PR. Both refs are available locally.

.github/workflows/rust-release.yml

@@ -181,9 +181,9 @@ jobs:
           name: ${{ steps.release_name.outputs.name }}
           tag_name: ${{ github.ref_name }}
           files: dist/**
-          # For now, tag releases as "prerelease" because we are not claiming
-          # the Rust CLI is stable yet.
-          prerelease: true
+          # Mark as prerelease only when the version has a suffix after x.y.z
+          # (e.g. -alpha, -beta). Otherwise publish a normal release.
+          prerelease: ${{ contains(steps.release_name.outputs.name, '-') }}
 
       - uses: facebook/dotslash-publish-release@v2
         env:

.vscode/settings.json

@@ -11,6 +11,8 @@
         "editor.defaultFormatter": "tamasfe.even-better-toml",
         "editor.formatOnSave": true,
     },
-    "evenBetterToml.formatter.reorderArrays": true,
+    // Array order for options in ~/.codex/config.toml such as `notify` and the
+    // `args` for an MCP server is significant, so we disable reordering.
+    "evenBetterToml.formatter.reorderArrays": false,
     "evenBetterToml.formatter.reorderKeys": true,
 }

README.md

@@ -18,6 +18,7 @@ This is the home of the **Codex CLI**, which is a coding agent from OpenAI that
 - [Quickstart](#quickstart)
   - [OpenAI API Users](#openai-api-users)
   - [OpenAI Plus/Pro Users](#openai-pluspro-users)
+- [Using Open Source Models](#using-open-source-models)
 - [Why Codex?](#why-codex)
 - [Security model & permissions](#security-model--permissions)
   - [Platform sandboxing details](#platform-sandboxing-details)
@@ -186,6 +187,41 @@ they'll be committed to your working directory.
 
 ---
 
+## Using Open Source Models
+
+Codex can run fully locally against an OpenAI-compatible OSS host (like Ollama) using the `--oss` flag:
+
+- Interactive UI:
+  - codex --oss
+- Non-interactive (programmatic) mode:
+  - echo "Refactor utils" | codex exec --oss
+
+Model selection when using `--oss`:
+
+- If you omit `-m/--model`, Codex defaults to -m gpt-oss:20b and will verify it exists locally (downloading if needed).
+- To pick a different size, pass one of:
+  - -m "gpt-oss:20b"
+  - -m "gpt-oss:120b"
+
+Point Codex at your own OSS host:
+
+- By default, `--oss` talks to http://localhost:11434/v1.
+- To use a different host, set one of these environment variables before running Codex:
+  - CODEX_OSS_BASE_URL, for example:
+    - CODEX_OSS_BASE_URL="http://my-ollama.example.com:11434/v1" codex --oss -m gpt-oss:20b
+  - or CODEX_OSS_PORT (when the host is localhost):
+    - CODEX_OSS_PORT=11434 codex --oss
+
+Advanced: you can persist this in your config instead of environment variables by overriding the built-in `oss` provider in `~/.codex/config.toml`:
+
+```toml
+[model_providers.oss]
+name = "Open Source"
+base_url = "http://my-ollama.example.com:11434/v1"
+```
+
+---
+
 ## Why Codex?
 
 Codex CLI is built for developers who already **live in the terminal** and want

codex-rs/Cargo.lock

@@ -729,6 +729,7 @@ dependencies = [
  "codex-arg0",
  "codex-common",
  "codex-core",
+ "codex-ollama",
  "owo-colors",
  "predicates",
  "serde_json",
@@ -838,6 +839,23 @@ dependencies = [
  "wiremock",
 ]
 
+[[package]]
+name = "codex-ollama"
+version = "0.0.0"
+dependencies = [
+ "async-stream",
+ "bytes",
+ "codex-core",
+ "futures",
+ "reqwest",
+ "serde_json",
+ "tempfile",
+ "tokio",
+ "toml 0.9.4",
+ "tracing",
+ "wiremock",
+]
+
 [[package]]
 name = "codex-tui"
 version = "0.0.0"
@@ -852,6 +870,7 @@ dependencies = [
  "codex-core",
  "codex-file-search",
  "codex-login",
+ "codex-ollama",
  "color-eyre",
  "crossterm",
  "image",

codex-rs/Cargo.toml

@@ -14,6 +14,7 @@ members = [
     "mcp-client",
     "mcp-server",
     "mcp-types",
+    "ollama",
     "tui",
 ]
 resolver = "2"

codex-rs/apply-patch/src/lib.rs

@@ -42,6 +42,15 @@ impl From<std::io::Error> for ApplyPatchError {
     }
 }
 
+impl From<&std::io::Error> for ApplyPatchError {
+    fn from(err: &std::io::Error) -> Self {
+        ApplyPatchError::IoError(IoError {
+            context: "I/O error".to_string(),
+            source: std::io::Error::new(err.kind(), err.to_string()),
+        })
+    }
+}
+
 #[derive(Debug, Error)]
 #[error("{context}: {source}")]
 pub struct IoError {
@@ -366,13 +375,21 @@ pub fn apply_hunks(
     match apply_hunks_to_files(hunks) {
         Ok(affected) => {
             print_summary(&affected, stdout).map_err(ApplyPatchError::from)?;
+            Ok(())
         }
         Err(err) => {
-            writeln!(stderr, "{err:?}").map_err(ApplyPatchError::from)?;
+            let msg = err.to_string();
+            writeln!(stderr, "{msg}").map_err(ApplyPatchError::from)?;
+            if let Some(io) = err.downcast_ref::<std::io::Error>() {
+                Err(ApplyPatchError::from(io))
+            } else {
+                Err(ApplyPatchError::IoError(IoError {
+                    context: msg,
+                    source: std::io::Error::other(err),
+                }))
+            }
         }
     }
-
-    Ok(())
 }
 
 /// Applies each parsed patch hunk to the filesystem.
@@ -1238,4 +1255,24 @@ g
             })
         );
     }
+
+    #[test]
+    fn test_apply_patch_fails_on_write_error() {
+        let dir = tempdir().unwrap();
+        let path = dir.path().join("readonly.txt");
+        fs::write(&path, "before\n").unwrap();
+        let mut perms = fs::metadata(&path).unwrap().permissions();
+        perms.set_readonly(true);
+        fs::set_permissions(&path, perms).unwrap();
+
+        let patch = wrap_patch(&format!(
+            "*** Update File: {}\n@@\n-before\n+after\n*** End Patch",
+            path.display()
+        ));
+
+        let mut stdout = Vec::new();
+        let mut stderr = Vec::new();
+        let result = apply_patch(&patch, &mut stdout, &mut stderr);
+        assert!(result.is_err());
+    }
 }

codex-rs/cli/src/main.rs

@@ -121,7 +121,9 @@ async fn cli_main(codex_linux_sandbox_exe: Option<PathBuf>) -> anyhow::Result<()
             let mut tui_cli = cli.interactive;
             prepend_config_flags(&mut tui_cli.config_overrides, cli.config_overrides);
             let usage = codex_tui::run_main(tui_cli, codex_linux_sandbox_exe).await?;
-            println!("{}", codex_core::protocol::FinalOutput::from(usage));
+            if !usage.is_zero() {
+                println!("{}", codex_core::protocol::FinalOutput::from(usage));
+            }
         }
         Some(Subcommand::Exec(mut exec_cli)) => {
             prepend_config_flags(&mut exec_cli.config_overrides, cli.config_overrides);

codex-rs/common/src/approval_mode_cli_arg.rs

@@ -18,6 +18,9 @@ pub enum ApprovalModeCliArg {
     /// will escalate to the user to ask for un-sandboxed execution.
     OnFailure,
 
+    /// The model decides when to ask the user for approval.
+    OnRequest,
+
     /// Never ask for user approval
     /// Execution failures are immediately returned to the model.
     Never,
@@ -28,6 +31,7 @@ impl From<ApprovalModeCliArg> for AskForApproval {
         match value {
             ApprovalModeCliArg::Untrusted => AskForApproval::UnlessTrusted,
             ApprovalModeCliArg::OnFailure => AskForApproval::OnFailure,
+            ApprovalModeCliArg::OnRequest => AskForApproval::OnRequest,
             ApprovalModeCliArg::Never => AskForApproval::Never,
         }
     }

codex-rs/common/src/config_summary.rs

@@ -0,0 +1,29 @@
+use codex_core::WireApi;
+use codex_core::config::Config;
+
+use crate::sandbox_summary::summarize_sandbox_policy;
+
+/// Build a list of key/value pairs summarizing the effective configuration.
+pub fn create_config_summary_entries(config: &Config) -> Vec<(&'static str, String)> {
+    let mut entries = vec![
+        ("workdir", config.cwd.display().to_string()),
+        ("model", config.model.clone()),
+        ("provider", config.model_provider_id.clone()),
+        ("approval", config.approval_policy.to_string()),
+        ("sandbox", summarize_sandbox_policy(&config.sandbox_policy)),
+    ];
+    if config.model_provider.wire_api == WireApi::Responses
+        && config.model_family.supports_reasoning_summaries
+    {
+        entries.push((
+            "reasoning effort",
+            config.model_reasoning_effort.to_string(),
+        ));
+        entries.push((
+            "reasoning summaries",
+            config.model_reasoning_summary.to_string(),
+        ));
+    }
+
+    entries
+}

codex-rs/common/src/lib.rs

@@ -23,3 +23,7 @@ mod sandbox_summary;
 
 #[cfg(feature = "sandbox_summary")]
 pub use sandbox_summary::summarize_sandbox_policy;
+
+mod config_summary;
+
+pub use config_summary::create_config_summary_entries;

codex-rs/config.md

@@ -148,12 +148,20 @@ Determines when the user should be prompted to approve whether Codex can execute
 approval_policy = "untrusted"
 ```
 
+If you want to be notified whenever a command fails, use "on-failure":
 ```toml
 # If the command fails when run in the sandbox, Codex asks for permission to
 # retry the command outside the sandbox.
 approval_policy = "on-failure"
 ```
 
+If you want the model to run until it decides that it needs to ask you for escalated permissions, use "on-request":
+```toml
+# The model decides when to escalate
+approval_policy = "on-request"
+```
+
+Alternatively, you can have the model run until it is done, and never ask to run a command with escalated permissions:
 ```toml
 # User is never prompted: if the command fails, Codex will automatically try
 # something out. Note the `exec` subcommand always uses this mode.
@@ -483,6 +491,19 @@ Setting `hide_agent_reasoning` to `true` suppresses these events in **both** the
 hide_agent_reasoning = true   # defaults to false
 ```
 
+## show_raw_agent_reasoning
+
+Surfaces the model’s raw chain-of-thought ("raw reasoning content") when available.
+
+Notes:
+- Only takes effect if the selected model/provider actually emits raw reasoning content. Many models do not. When unsupported, this option has no visible effect.
+- Raw reasoning may include intermediate thoughts or sensitive context. Enable only if acceptable for your workflow.
+
+Example:
+```toml
+show_raw_agent_reasoning = true  # defaults to false
+```
+
 ## model_context_window
 
 The size of the context window for the model, in tokens.

codex-rs/core/prompt.md

@@ -1,8 +1,21 @@
-Please resolve the user's task by editing and testing the code files in your current code execution session.
-You are a deployed coding agent.
-Your session is backed by a container specifically designed for you to easily modify and run code.
-The repo(s) are already cloned in your working directory, and you must fully solve the problem for your answer to be considered correct.
+You are operating as and within the Codex CLI, an open-source, terminal-based agentic coding assistant built by OpenAI. It wraps OpenAI models to enable natural language interaction with a local codebase. You are expected to be precise, safe, and helpful.
 
+Your capabilities:
+- Receive user prompts, project context, and files.
+- Stream responses and emit function calls (e.g., shell commands, code edits).
+- Run commands, like apply_patch, and manage user approvals based on policy.
+- Work inside a workspace with sandboxing instructions specified by the policy described in (## Sandbox environment and approval instructions)
+
+Within this context, Codex refers to the open-source agentic coding interface (not the old Codex language model built by OpenAI).
+
+## General guidelines
+As a deployed coding agent, please continue working on the user's task until their query is resolved, before ending your turn and yielding back to the user. Only terminate your turn when you are sure that the task is solved. If you are not sure about file content or codebase structure pertaining to the user's request, use your tools to read files and gather the relevant information. Do NOT guess or make up an answer.
+
+After a user sends their first message, you should immediately provide a brief message acknowledging their request to set the tone and expectation of future work to be done (no more than 8-10 words). This should be done before performing work like exploring the codebase, writing or reading files, or other tool calls needed to complete the task. Use a natural, collaborative tone similar to how a teammate would receive a task during a pair programming session.
+
+Please resolve the user's task by editing the code files in your current code execution session. Your session allows for you to modify and run code. The repo(s) are already cloned in your working directory, and you must fully solve the problem for your answer to be considered correct.
+
+### Task execution
 You MUST adhere to the following criteria when executing the task:
 
 - Working on the repo(s) in the current environment is allowed, even if they are proprietary.
@@ -12,7 +25,7 @@ You MUST adhere to the following criteria when executing the task:
 - `user_instructions` are not part of the user's request, but guidance for how to complete the task.
 - Do not cite `user_instructions` back to the user unless a specific piece is relevant.
 - Do not use \`ls -R\`, \`find\`, or \`grep\` - these are slow in large repos. Use \`rg\` and \`rg --files\`.
-- Use \`apply_patch\` to edit files: {"command":["apply_patch","*** Begin Patch\\n*** Update File: path/to/file.py\\n@@ def example():\\n- pass\\n+ return 123\\n*** End Patch"]}
+- Use the \`apply_patch\` shell command to edit files: {"command":["apply_patch","*** Begin Patch\\n*** Update File: path/to/file.py\\n@@ def example():\\n- pass\\n+ return 123\\n*** End Patch"]}
 - If completing the user's task requires writing or modifying files:
   - Your code and final answer should follow these _CODING GUIDELINES_:
     - Fix the problem at the root cause rather than applying surface-level patches, when possible.
@@ -35,12 +48,11 @@ You MUST adhere to the following criteria when executing the task:
 - If completing the user's task DOES NOT require writing or modifying files (e.g., the user asks a question about the code base):
   - Respond in a friendly tune as a remote teammate, who is knowledgeable, capable and eager to help with coding.
 - When your task involves writing or modifying files:
-  - Do NOT tell the user to "save the file" or "copy the code into a file" if you already created or modified the file using \`apply_patch\`. Instead, reference the file as already saved.
+  - Do NOT tell the user to "save the file" or "copy the code into a file" if you already created or modified the file using the `apply_patch` shell command. Instead, reference the file as already saved.
   - Do NOT show the full contents of large files you have already written, unless the user explicitly asks for them.
 
-§ `apply-patch` Specification
-
-Your patch language is a stripped‑down, file‑oriented diff format designed to be easy to parse and safe to apply. You can think of it as a high‑level envelope:
+## Using the shell command `apply_patch` to edit files
+`apply_patch` is a shell command for editing files. Your patch language is a stripped‑down, file‑oriented diff format designed to be easy to parse and safe to apply. You can think of it as a high‑level envelope:
 
 *** Begin Patch
 [ one or more file sections ]
@@ -92,14 +104,28 @@ It is important to remember:
 
 - You must include a header with your intended action (Add/Delete/Update)
 - You must prefix new lines with `+` even when creating a new file
+- You must follow this schema exactly when providing a patch
 
-You can invoke apply_patch like:
+You can invoke apply_patch with the following shell command:
 
 ```
 shell {"command":["apply_patch","*** Begin Patch\n*** Add File: hello.txt\n+Hello, world!\n*** End Patch\n"]}
 ```
 
-Plan updates
+## Sandbox environment and approval instructions
+
+You are running in a sandboxed workspace backed by version control. The sandbox might be configured by the user to restrict certain behaviors, like accessing the internet or writing to files outside the current directory.
+
+Commands that are blocked by sandbox settings will be automatically sent to the user for approval. The result of the request will be returned (i.e. the command result, or the request denial).
+The user also has an opportunity to approve the same command for the rest of the session.
+
+Guidance on running within the sandbox:
+- When running commands that will likely require approval, attempt to use simple, precise commands, to reduce frequency of approval requests.
+- When approval is denied or a command fails due to a permission error, do not retry the exact command in a different way. Move on and continue trying to address the user's request.
+
+
+## Tools available
+### Plan updates
 
 A tool named `update_plan` is available. Use it to keep an up‑to‑date, step‑by‑step plan for the task so you can follow your progress. When making your plans, keep in mind that you are a deployed coding agent - `update_plan` calls should not involve doing anything that you aren't capable of doing. For example, `update_plan` calls should NEVER contain tasks to merge your own pull requests. Only stop to ask the user if you genuinely need their feedback on a change.
 
@@ -107,3 +133,4 @@ A tool named `update_plan` is available. Use it to keep an up‑to‑date, step
 - Whenever you finish a step, call `update_plan` again, marking the finished step as `completed` and the next step as `in_progress`.
 - If your plan needs to change, call `update_plan` with the revised steps and include an `explanation` describing the change.
 - When all steps are complete, make a final `update_plan` call with all steps marked `completed`.
+

codex-rs/core/src/apply_patch.rs

@@ -1,6 +1,9 @@
 use crate::codex::Session;
 use crate::models::FunctionCallOutputPayload;
 use crate::models::ResponseInputItem;
+use crate::protocol::DiffHunk;
+use crate::protocol::DiffLine;
+use crate::protocol::DiffLineKind;
 use crate::protocol::FileChange;
 use crate::protocol::ReviewDecision;
 use crate::safety::SafetyCheck;
@@ -119,16 +122,143 @@ pub(crate) fn convert_apply_patch_to_protocol(
                 unified_diff,
                 move_path,
                 new_content: _new_content,
-            } => FileChange::Update {
-                unified_diff: unified_diff.clone(),
-                move_path: move_path.clone(),
-            },
+            } => {
+                let hunks = parse_unified_diff_to_hunks(unified_diff).unwrap_or_default();
+                FileChange::Update {
+                    unified_diff: unified_diff.clone(),
+                    move_path: move_path.clone(),
+                    hunks: if hunks.is_empty() { None } else { Some(hunks) },
+                }
+            }
         };
         result.insert(path.clone(), protocol_change);
     }
     result
 }
 
+/// Parse a unified diff string into structured hunks. The input is expected to
+/// contain one or more hunk headers (lines starting with "@@") followed by hunk
+/// bodies. Lines starting with '+++', '---' (file headers) are ignored.
+fn parse_unified_diff_to_hunks(src: &str) -> Option<Vec<DiffHunk>> {
+    let mut hunks: Vec<DiffHunk> = Vec::new();
+    let mut cur: Option<DiffHunk> = None;
+
+    for line in src.lines() {
+        if line.starts_with("---") || line.starts_with("+++") {
+            // File headers: ignore.
+            continue;
+        }
+
+        if let Some((old_start, old_count, new_start, new_count)) = parse_hunk_header(line) {
+            // Flush previous hunk
+            if let Some(h) = cur.take() {
+                hunks.push(h);
+            }
+            cur = Some(DiffHunk {
+                old_start: old_start as u32,
+                old_count: old_count as u32,
+                new_start: new_start as u32,
+                new_count: new_count as u32,
+                lines: Vec::new(),
+            });
+            continue;
+        }
+
+        if let Some(h) = cur.as_mut() {
+            // Classify by prefix; store text without the prefix when present.
+            let (kind, text) = if let Some(rest) = line.strip_prefix('+') {
+                (DiffLineKind::Add, rest.to_string())
+            } else if let Some(rest) = line.strip_prefix('-') {
+                (DiffLineKind::Delete, rest.to_string())
+            } else if let Some(rest) = line.strip_prefix(' ') {
+                (DiffLineKind::Context, rest.to_string())
+            } else {
+                // Non-standard line inside hunk; keep as context with full text.
+                (DiffLineKind::Context, line.to_string())
+            };
+            h.lines.push(DiffLine { kind, text });
+        }
+    }
+
+    if let Some(h) = cur.take() {
+        hunks.push(h);
+    }
+
+    Some(hunks)
+}
+
+// Lightweight parsing of a unified diff hunk header of the form:
+// @@ -oldStart,oldCount +newStart,newCount @@
+// Counts may be omitted which implies 1.
+fn parse_hunk_header(line: &str) -> Option<(u64, u64, u64, u64)> {
+    if !line.starts_with("@@") {
+        return None;
+    }
+    let bytes = line.as_bytes();
+    let mut i = 2usize;
+    // skip spaces
+    while i < bytes.len() && bytes[i].is_ascii_whitespace() {
+        i += 1;
+    }
+    if i >= bytes.len() || bytes[i] != b'-' {
+        return None;
+    }
+    i += 1;
+    let (old_start, c1) = parse_uint(&bytes[i..]);
+    if c1 == 0 {
+        return None;
+    }
+    i += c1;
+    let mut old_count = 1u64;
+    if i < bytes.len() && bytes[i] == b',' {
+        i += 1;
+        let (n, c) = parse_uint(&bytes[i..]);
+        if c == 0 {
+            return None;
+        }
+        old_count = n;
+        i += c;
+    }
+    while i < bytes.len() && bytes[i].is_ascii_whitespace() {
+        i += 1;
+    }
+    if i >= bytes.len() || bytes[i] != b'+' {
+        return None;
+    }
+    i += 1;
+    let (new_start, c2) = parse_uint(&bytes[i..]);
+    if c2 == 0 {
+        return None;
+    }
+    i += c2;
+    let mut new_count = 1u64;
+    if i < bytes.len() && bytes[i] == b',' {
+        i += 1;
+        let (n, c) = parse_uint(&bytes[i..]);
+        if c == 0 {
+            return None;
+        }
+        new_count = n;
+        i += c;
+    }
+    Some((old_start, old_count, new_start, new_count))
+}
+
+fn parse_uint(s: &[u8]) -> (u64, usize) {
+    let mut i = 0usize;
+    let mut n: u64 = 0;
+    while i < s.len() {
+        let b = s[i];
+        if b.is_ascii_digit() {
+            n = n * 10 + (b - b'0') as u64;
+            i += 1;
+        } else {
+            break;
+        }
+    }
+    (n, i)
+}
+
 pub(crate) fn get_writable_roots(cwd: &Path) -> Vec<PathBuf> {
     let mut writable_roots = Vec::new();
     if cfg!(target_os = "macos") {

codex-rs/core/src/chat_completions.rs

@@ -21,7 +21,9 @@ use crate::client_common::ResponseEvent;
 use crate::client_common::ResponseStream;
 use crate::error::CodexErr;
 use crate::error::Result;
+use crate::model_family::ModelFamily;
 use crate::models::ContentItem;
+use crate::models::ReasoningItemContent;
 use crate::models::ResponseItem;
 use crate::openai_tools::create_tools_json_for_chat_completions_api;
 use crate::util::backoff;
@@ -29,22 +31,19 @@ use crate::util::backoff;
 /// Implementation for the classic Chat Completions API.
 pub(crate) async fn stream_chat_completions(
     prompt: &Prompt,
-    model: &str,
-    include_plan_tool: bool,
+    model_family: &ModelFamily,
     client: &reqwest::Client,
     provider: &ModelProviderInfo,
 ) -> Result<ResponseStream> {
     // Build messages array
     let mut messages = Vec::<serde_json::Value>::new();
 
-    let full_instructions = prompt.get_full_instructions(model);
+    let full_instructions = prompt.get_full_instructions(model_family);
     messages.push(json!({"role": "system", "content": full_instructions}));
 
-    if let Some(instr) = &prompt.get_formatted_user_instructions() {
-        messages.push(json!({"role": "user", "content": instr}));
-    }
+    let input = prompt.get_formatted_input();
 
-    for item in &prompt.input {
+    for item in &input {
         match item {
             ResponseItem::Message { role, content, .. } => {
                 let mut text = String::new();
@@ -110,9 +109,9 @@ pub(crate) async fn stream_chat_completions(
         }
     }
 
-    let tools_json = create_tools_json_for_chat_completions_api(prompt, model, include_plan_tool)?;
+    let tools_json = create_tools_json_for_chat_completions_api(&prompt.tools)?;
     let payload = json!({
-        "model": model,
+        "model": model_family.slug,
         "messages": messages,
         "stream": true,
         "tools": tools_json,
@@ -207,6 +206,8 @@ async fn process_chat_sse<S>(
     }
 
     let mut fn_call_state = FunctionCallState::default();
+    let mut assistant_text = String::new();
+    let mut reasoning_text = String::new();
 
     loop {
         let sse = match timeout(idle_timeout, stream.next()).await {
@@ -235,6 +236,31 @@ async fn process_chat_sse<S>(
 
         // OpenAI Chat streaming sends a literal string "[DONE]" when finished.
         if sse.data.trim() == "[DONE]" {
+            // Emit any finalized items before closing so downstream consumers receive
+            // terminal events for both assistant content and raw reasoning.
+            if !assistant_text.is_empty() {
+                let item = ResponseItem::Message {
+                    role: "assistant".to_string(),
+                    content: vec![ContentItem::OutputText {
+                        text: std::mem::take(&mut assistant_text),
+                    }],
+                    id: None,
+                };
+                let _ = tx_event.send(Ok(ResponseEvent::OutputItemDone(item))).await;
+            }
+
+            if !reasoning_text.is_empty() {
+                let item = ResponseItem::Reasoning {
+                    id: String::new(),
+                    summary: Vec::new(),
+                    content: Some(vec![ReasoningItemContent::ReasoningText {
+                        text: std::mem::take(&mut reasoning_text),
+                    }]),
+                    encrypted_content: None,
+                };
+                let _ = tx_event.send(Ok(ResponseEvent::OutputItemDone(item))).await;
+            }
+
             let _ = tx_event
                 .send(Ok(ResponseEvent::Completed {
                     response_id: String::new(),
@@ -254,26 +280,47 @@ async fn process_chat_sse<S>(
         let choice_opt = chunk.get("choices").and_then(|c| c.get(0));
 
         if let Some(choice) = choice_opt {
-            // Handle assistant content tokens.
+            // Handle assistant content tokens as streaming deltas.
             if let Some(content) = choice
                 .get("delta")
                 .and_then(|d| d.get("content"))
                 .and_then(|c| c.as_str())
             {
-                // Emit a delta so downstream consumers can stream text live.
-                let _ = tx_event
-                    .send(Ok(ResponseEvent::OutputTextDelta(content.to_string())))
-                    .await;
+                if !content.is_empty() {
+                    assistant_text.push_str(content);
+                    let _ = tx_event
+                        .send(Ok(ResponseEvent::OutputTextDelta(content.to_string())))
+                        .await;
+                }
+            }
 
-                let item = ResponseItem::Message {
-                    role: "assistant".to_string(),
-                    content: vec![ContentItem::OutputText {
-                        text: content.to_string(),
-                    }],
-                    id: None,
-                };
+            // Forward any reasoning/thinking deltas if present.
+            // Some providers stream `reasoning` as a plain string while others
+            // nest the text under an object (e.g. `{ "reasoning": { "text": "…" } }`).
+            if let Some(reasoning_val) = choice.get("delta").and_then(|d| d.get("reasoning")) {
+                let mut maybe_text = reasoning_val.as_str().map(|s| s.to_string());
 
-                let _ = tx_event.send(Ok(ResponseEvent::OutputItemDone(item))).await;
+                if maybe_text.is_none() && reasoning_val.is_object() {
+                    if let Some(s) = reasoning_val
+                        .get("text")
+                        .and_then(|t| t.as_str())
+                        .filter(|s| !s.is_empty())
+                    {
+                        maybe_text = Some(s.to_string());
+                    } else if let Some(s) = reasoning_val
+                        .get("content")
+                        .and_then(|t| t.as_str())
+                        .filter(|s| !s.is_empty())
+                    {
+                        maybe_text = Some(s.to_string());
+                    }
+                }
+
+                if let Some(reasoning) = maybe_text {
+                    let _ = tx_event
+                        .send(Ok(ResponseEvent::ReasoningContentDelta(reasoning)))
+                        .await;
+                }
             }
 
             // Handle streaming function / tool calls.
@@ -310,7 +357,21 @@ async fn process_chat_sse<S>(
             if let Some(finish_reason) = choice.get("finish_reason").and_then(|v| v.as_str()) {
                 match finish_reason {
                     "tool_calls" if fn_call_state.active => {
-                        // Build the FunctionCall response item.
+                        // First, flush the terminal raw reasoning so UIs can finalize
+                        // the reasoning stream before any exec/tool events begin.
+                        if !reasoning_text.is_empty() {
+                            let item = ResponseItem::Reasoning {
+                                id: String::new(),
+                                summary: Vec::new(),
+                                content: Some(vec![ReasoningItemContent::ReasoningText {
+                                    text: std::mem::take(&mut reasoning_text),
+                                }]),
+                                encrypted_content: None,
+                            };
+                            let _ = tx_event.send(Ok(ResponseEvent::OutputItemDone(item))).await;
+                        }
+
+                        // Then emit the FunctionCall response item.
                         let item = ResponseItem::FunctionCall {
                             id: None,
                             name: fn_call_state.name.clone().unwrap_or_else(|| "".to_string()),
@@ -318,11 +379,33 @@ async fn process_chat_sse<S>(
                             call_id: fn_call_state.call_id.clone().unwrap_or_else(String::new),
                         };
 
-                        // Emit it downstream.
                         let _ = tx_event.send(Ok(ResponseEvent::OutputItemDone(item))).await;
                     }
                     "stop" => {
-                        // Regular turn without tool-call.
+                        // Regular turn without tool-call. Emit the final assistant message
+                        // as a single OutputItemDone so non-delta consumers see the result.
+                        if !assistant_text.is_empty() {
+                            let item = ResponseItem::Message {
+                                role: "assistant".to_string(),
+                                content: vec![ContentItem::OutputText {
+                                    text: std::mem::take(&mut assistant_text),
+                                }],
+                                id: None,
+                            };
+                            let _ = tx_event.send(Ok(ResponseEvent::OutputItemDone(item))).await;
+                        }
+                        // Also emit a terminal Reasoning item so UIs can finalize raw reasoning.
+                        if !reasoning_text.is_empty() {
+                            let item = ResponseItem::Reasoning {
+                                id: String::new(),
+                                summary: Vec::new(),
+                                content: Some(vec![ReasoningItemContent::ReasoningText {
+                                    text: std::mem::take(&mut reasoning_text),
+                                }]),
+                                encrypted_content: None,
+                            };
+                            let _ = tx_event.send(Ok(ResponseEvent::OutputItemDone(item))).await;
+                        }
                     }
                     _ => {}
                 }
@@ -360,10 +443,17 @@ async fn process_chat_sse<S>(
 /// The adapter is intentionally *lossless*: callers who do **not** opt in via
 /// [`AggregateStreamExt::aggregate()`] keep receiving the original unmodified
 /// events.
+#[derive(Copy, Clone, Eq, PartialEq)]
+enum AggregateMode {
+    AggregatedOnly,
+    Streaming,
+}
 pub(crate) struct AggregatedChatStream<S> {
     inner: S,
     cumulative: String,
-    pending_completed: Option<ResponseEvent>,
+    cumulative_reasoning: String,
+    pending: std::collections::VecDeque<ResponseEvent>,
+    mode: AggregateMode,
 }
 
 impl<S> Stream for AggregatedChatStream<S>
@@ -375,8 +465,8 @@ where
     fn poll_next(self: Pin<&mut Self>, cx: &mut Context<'_>) -> Poll<Option<Self::Item>> {
         let this = self.get_mut();
 
-        // First, flush any buffered Completed event from the previous call.
-        if let Some(ev) = this.pending_completed.take() {
+        // First, flush any buffered events from the previous call.
+        if let Some(ev) = this.pending.pop_front() {
             return Poll::Ready(Some(Ok(ev)));
         }
 
@@ -393,16 +483,21 @@ where
                     let is_assistant_delta = matches!(&item, crate::models::ResponseItem::Message { role, .. } if role == "assistant");
 
                     if is_assistant_delta {
-                        if let crate::models::ResponseItem::Message { content, .. } = &item {
-                            if let Some(text) = content.iter().find_map(|c| match c {
-                                crate::models::ContentItem::OutputText { text } => Some(text),
-                                _ => None,
-                            }) {
-                                this.cumulative.push_str(text);
+                        // Only use the final assistant message if we have not
+                        // seen any deltas; otherwise, deltas already built the
+                        // cumulative text and this would duplicate it.
+                        if this.cumulative.is_empty() {
+                            if let crate::models::ResponseItem::Message { content, .. } = &item {
+                                if let Some(text) = content.iter().find_map(|c| match c {
+                                    crate::models::ContentItem::OutputText { text } => Some(text),
+                                    _ => None,
+                                }) {
+                                    this.cumulative.push_str(text);
+                                }
                             }
                         }
 
-                        // Swallow partial assistant chunk; keep polling.
+                        // Swallow assistant message here; emit on Completed.
                         continue;
                     }
 
@@ -413,24 +508,50 @@ where
                     response_id,
                     token_usage,
                 }))) => {
+                    // Build any aggregated items in the correct order: Reasoning first, then Message.
+                    let mut emitted_any = false;
+
+                    if !this.cumulative_reasoning.is_empty()
+                        && matches!(this.mode, AggregateMode::AggregatedOnly)
+                    {
+                        let aggregated_reasoning = crate::models::ResponseItem::Reasoning {
+                            id: String::new(),
+                            summary: Vec::new(),
+                            content: Some(vec![
+                                crate::models::ReasoningItemContent::ReasoningText {
+                                    text: std::mem::take(&mut this.cumulative_reasoning),
+                                },
+                            ]),
+                            encrypted_content: None,
+                        };
+                        this.pending
+                            .push_back(ResponseEvent::OutputItemDone(aggregated_reasoning));
+                        emitted_any = true;
+                    }
+
                     if !this.cumulative.is_empty() {
-                        let aggregated_item = crate::models::ResponseItem::Message {
+                        let aggregated_message = crate::models::ResponseItem::Message {
                             id: None,
                             role: "assistant".to_string(),
                             content: vec![crate::models::ContentItem::OutputText {
                                 text: std::mem::take(&mut this.cumulative),
                             }],
                         };
+                        this.pending
+                            .push_back(ResponseEvent::OutputItemDone(aggregated_message));
+                        emitted_any = true;
+                    }
 
-                        // Buffer Completed so it is returned *after* the aggregated message.
-                        this.pending_completed = Some(ResponseEvent::Completed {
-                            response_id,
-                            token_usage,
+                    // Always emit Completed last when anything was aggregated.
+                    if emitted_any {
+                        this.pending.push_back(ResponseEvent::Completed {
+                            response_id: response_id.clone(),
+                            token_usage: token_usage.clone(),
                         });
-
-                        return Poll::Ready(Some(Ok(ResponseEvent::OutputItemDone(
-                            aggregated_item,
-                        ))));
+                        // Return the first pending event now.
+                        if let Some(ev) = this.pending.pop_front() {
+                            return Poll::Ready(Some(Ok(ev)));
+                        }
                     }
 
                     // Nothing aggregated – forward Completed directly.
@@ -445,13 +566,27 @@ where
                     continue;
                 }
                 Poll::Ready(Some(Ok(ResponseEvent::OutputTextDelta(delta)))) => {
-                    // Forward deltas unchanged so callers can stream text
-                    // live while still receiving a single aggregated
-                    // OutputItemDone at the end of the turn.
-                    return Poll::Ready(Some(Ok(ResponseEvent::OutputTextDelta(delta))));
+                    // Always accumulate deltas so we can emit a final OutputItemDone at Completed.
+                    this.cumulative.push_str(&delta);
+                    if matches!(this.mode, AggregateMode::Streaming) {
+                        // In streaming mode, also forward the delta immediately.
+                        return Poll::Ready(Some(Ok(ResponseEvent::OutputTextDelta(delta))));
+                    } else {
+                        continue;
+                    }
                 }
-                Poll::Ready(Some(Ok(ResponseEvent::ReasoningSummaryDelta(delta)))) => {
-                    return Poll::Ready(Some(Ok(ResponseEvent::ReasoningSummaryDelta(delta))));
+                Poll::Ready(Some(Ok(ResponseEvent::ReasoningContentDelta(delta)))) => {
+                    // Always accumulate reasoning deltas so we can emit a final Reasoning item at Completed.
+                    this.cumulative_reasoning.push_str(&delta);
+                    if matches!(this.mode, AggregateMode::Streaming) {
+                        // In streaming mode, also forward the delta immediately.
+                        return Poll::Ready(Some(Ok(ResponseEvent::ReasoningContentDelta(delta))));
+                    } else {
+                        continue;
+                    }
+                }
+                Poll::Ready(Some(Ok(ResponseEvent::ReasoningSummaryDelta(_)))) => {
+                    continue;
                 }
             }
         }
@@ -480,12 +615,24 @@ pub(crate) trait AggregateStreamExt: Stream<Item = Result<ResponseEvent>> + Size
     /// }
     /// ```
     fn aggregate(self) -> AggregatedChatStream<Self> {
-        AggregatedChatStream {
-            inner: self,
-            cumulative: String::new(),
-            pending_completed: None,
-        }
+        AggregatedChatStream::new(self, AggregateMode::AggregatedOnly)
     }
 }
 
 impl<T> AggregateStreamExt for T where T: Stream<Item = Result<ResponseEvent>> + Sized {}
+
+impl<S> AggregatedChatStream<S> {
+    fn new(inner: S, mode: AggregateMode) -> Self {
+        AggregatedChatStream {
+            inner,
+            cumulative: String::new(),
+            cumulative_reasoning: String::new(),
+            pending: std::collections::VecDeque::new(),
+            mode,
+        }
+    }
+
+    pub(crate) fn streaming_mode(inner: S) -> Self {
+        Self::new(inner, AggregateMode::Streaming)
+    }
+}

codex-rs/core/src/client.rs

@@ -34,7 +34,6 @@ use crate::error::Result;
 use crate::flags::CODEX_RS_SSE_FIXTURE;
 use crate::model_provider_info::ModelProviderInfo;
 use crate::model_provider_info::WireApi;
-use crate::models::ContentItem;
 use crate::models::ResponseItem;
 use crate::openai_tools::create_tools_json_for_responses_api;
 use crate::protocol::TokenUsage;
@@ -82,8 +81,7 @@ impl ModelClient {
                 // Create the raw streaming connection first.
                 let response_stream = stream_chat_completions(
                     prompt,
-                    &self.config.model,
-                    self.config.include_plan_tool,
+                    &self.config.model_family,
                     &self.client,
                     &self.provider,
                 )
@@ -92,7 +90,11 @@ impl ModelClient {
                 // Wrap it with the aggregation adapter so callers see *only*
                 // the final assistant message per turn (matching the
                 // behaviour of the Responses API).
-                let mut aggregated = response_stream.aggregate();
+                let mut aggregated = if self.config.show_raw_agent_reasoning {
+                    crate::chat_completions::AggregatedChatStream::streaming_mode(response_stream)
+                } else {
+                    response_stream.aggregate()
+                };
 
                 // Bridge the aggregated stream back into a standard
                 // `ResponseStream` by forwarding events through a channel.
@@ -125,15 +127,24 @@ impl ModelClient {
 
         let auth_mode = auth.as_ref().map(|a| a.mode);
 
+        if self.config.model_family.family == "2025-08-06-model"
+            && auth_mode != Some(AuthMode::ChatGPT)
+        {
+            return Err(CodexErr::UnexpectedStatus(
+                StatusCode::BAD_REQUEST,
+                "2025-08-06-model is only supported with ChatGPT auth, run `codex login status` to check your auth status and `codex login` to login with ChatGPT".to_string(),
+            ));
+        }
+
         let store = prompt.store && auth_mode != Some(AuthMode::ChatGPT);
 
-        let full_instructions = prompt.get_full_instructions(&self.config.model);
-        let tools_json = create_tools_json_for_responses_api(
-            prompt,
-            &self.config.model,
-            self.config.include_plan_tool,
-        )?;
-        let reasoning = create_reasoning_param_for_request(&self.config, self.effort, self.summary);
+        let full_instructions = prompt.get_full_instructions(&self.config.model_family);
+        let tools_json = create_tools_json_for_responses_api(&prompt.tools)?;
+        let reasoning = create_reasoning_param_for_request(
+            &self.config.model_family,
+            self.effort,
+            self.summary,
+        );
 
         // Request encrypted COT if we are not storing responses,
         // otherwise reasoning items will be referenced by ID
@@ -143,15 +154,7 @@ impl ModelClient {
             vec![]
         };
 
-        let mut input_with_instructions = Vec::with_capacity(prompt.input.len() + 1);
-        if let Some(ui) = prompt.get_formatted_user_instructions() {
-            input_with_instructions.push(ResponseItem::Message {
-                id: None,
-                role: "user".to_string(),
-                content: vec![ContentItem::InputText { text: ui }],
-            });
-        }
-        input_with_instructions.extend(prompt.input.clone());
+        let input_with_instructions = prompt.get_formatted_input();
 
         let payload = ResponsesApiRequest {
             model: &self.config.model,
@@ -433,6 +436,14 @@ async fn process_sse<S>(
                     }
                 }
             }
+            "response.reasoning_text.delta" => {
+                if let Some(delta) = event.delta {
+                    let event = ResponseEvent::ReasoningContentDelta(delta);
+                    if tx_event.send(Ok(event)).await.is_err() {
+                        return;
+                    }
+                }
+            }
             "response.created" => {
                 if event.response.is_some() {
                     let _ = tx_event.send(Ok(ResponseEvent::Created {})).await;
@@ -621,7 +632,7 @@ mod tests {
             request_max_retries: Some(0),
             stream_max_retries: Some(0),
             stream_idle_timeout_ms: Some(1000),
-            requires_auth: false,
+            requires_openai_auth: false,
         };
 
         let events = collect_events(
@@ -681,7 +692,7 @@ mod tests {
             request_max_retries: Some(0),
             stream_max_retries: Some(0),
             stream_idle_timeout_ms: Some(1000),
-            requires_auth: false,
+            requires_openai_auth: false,
         };
 
         let events = collect_events(&[sse1.as_bytes()], provider).await;
@@ -784,7 +795,7 @@ mod tests {
                 request_max_retries: Some(0),
                 stream_max_retries: Some(0),
                 stream_idle_timeout_ms: Some(1000),
-                requires_auth: false,
+                requires_openai_auth: false,
             };
 
             let out = run_sse(evs, provider).await;

codex-rs/core/src/client_common.rs

@@ -1,13 +1,19 @@
 use crate::config_types::ReasoningEffort as ReasoningEffortConfig;
 use crate::config_types::ReasoningSummary as ReasoningSummaryConfig;
 use crate::error::Result;
+use crate::model_family::ModelFamily;
+use crate::models::ContentItem;
 use crate::models::ResponseItem;
+use crate::openai_tools::OpenAiTool;
+use crate::protocol::AskForApproval;
+use crate::protocol::SandboxPolicy;
 use crate::protocol::TokenUsage;
 use codex_apply_patch::APPLY_PATCH_TOOL_INSTRUCTIONS;
 use futures::Stream;
 use serde::Serialize;
 use std::borrow::Cow;
-use std::collections::HashMap;
+use std::fmt::Display;
+use std::path::PathBuf;
 use std::pin::Pin;
 use std::task::Context;
 use std::task::Poll;
@@ -17,10 +23,47 @@ use tokio::sync::mpsc;
 /// with this content.
 const BASE_INSTRUCTIONS: &str = include_str!("../prompt.md");
 
+/// wraps environment context message in a tag for the model to parse more easily.
+const ENVIRONMENT_CONTEXT_START: &str = "<environment_context>\n\n";
+const ENVIRONMENT_CONTEXT_END: &str = "\n\n</environment_context>";
+
 /// wraps user instructions message in a tag for the model to parse more easily.
 const USER_INSTRUCTIONS_START: &str = "<user_instructions>\n\n";
 const USER_INSTRUCTIONS_END: &str = "\n\n</user_instructions>";
 
+#[derive(Debug, Clone)]
+pub(crate) struct EnvironmentContext {
+    pub cwd: PathBuf,
+    pub approval_policy: AskForApproval,
+    pub sandbox_policy: SandboxPolicy,
+}
+
+impl Display for EnvironmentContext {
+    fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
+        writeln!(
+            f,
+            "Current working directory: {}",
+            self.cwd.to_string_lossy()
+        )?;
+        writeln!(f, "Approval policy: {}", self.approval_policy)?;
+        writeln!(f, "Sandbox policy: {}", self.sandbox_policy)?;
+
+        let network_access = match self.sandbox_policy.clone() {
+            SandboxPolicy::DangerFullAccess => "enabled",
+            SandboxPolicy::ReadOnly => "restricted",
+            SandboxPolicy::WorkspaceWrite { network_access, .. } => {
+                if network_access {
+                    "enabled"
+                } else {
+                    "restricted"
+                }
+            }
+        };
+        writeln!(f, "Network access: {network_access}")?;
+        Ok(())
+    }
+}
+
 /// API request payload for a single model turn.
 #[derive(Default, Debug, Clone)]
 pub struct Prompt {
@@ -32,33 +75,62 @@ pub struct Prompt {
     /// Whether to store response on server side (disable_response_storage = !store).
     pub store: bool,
 
-    /// Additional tools sourced from external MCP servers. Note each key is
-    /// the "fully qualified" tool name (i.e., prefixed with the server name),
-    /// which should be reported to the model in place of Tool::name.
-    pub extra_tools: HashMap<String, mcp_types::Tool>,
+    /// A list of key-value pairs that will be added as a developer message
+    /// for the model to use
+    pub environment_context: Option<EnvironmentContext>,
+
+    /// Tools available to the model, including additional tools sourced from
+    /// external MCP servers.
+    pub tools: Vec<OpenAiTool>,
 
     /// Optional override for the built-in BASE_INSTRUCTIONS.
     pub base_instructions_override: Option<String>,
 }
 
 impl Prompt {
-    pub(crate) fn get_full_instructions(&self, model: &str) -> Cow<'_, str> {
+    pub(crate) fn get_full_instructions(&self, model: &ModelFamily) -> Cow<'_, str> {
         let base = self
             .base_instructions_override
             .as_deref()
             .unwrap_or(BASE_INSTRUCTIONS);
         let mut sections: Vec<&str> = vec![base];
-        if model.starts_with("gpt-4.1") {
+        if model.needs_special_apply_patch_instructions {
             sections.push(APPLY_PATCH_TOOL_INSTRUCTIONS);
         }
         Cow::Owned(sections.join("\n"))
     }
 
-    pub(crate) fn get_formatted_user_instructions(&self) -> Option<String> {
+    fn get_formatted_user_instructions(&self) -> Option<String> {
         self.user_instructions
             .as_ref()
             .map(|ui| format!("{USER_INSTRUCTIONS_START}{ui}{USER_INSTRUCTIONS_END}"))
     }
+
+    fn get_formatted_environment_context(&self) -> Option<String> {
+        self.environment_context
+            .as_ref()
+            .map(|ec| format!("{ENVIRONMENT_CONTEXT_START}{ec}{ENVIRONMENT_CONTEXT_END}"))
+    }
+
+    pub(crate) fn get_formatted_input(&self) -> Vec<ResponseItem> {
+        let mut input_with_instructions = Vec::with_capacity(self.input.len() + 2);
+        if let Some(ec) = self.get_formatted_environment_context() {
+            input_with_instructions.push(ResponseItem::Message {
+                id: None,
+                role: "user".to_string(),
+                content: vec![ContentItem::InputText { text: ec }],
+            });
+        }
+        if let Some(ui) = self.get_formatted_user_instructions() {
+            input_with_instructions.push(ResponseItem::Message {
+                id: None,
+                role: "user".to_string(),
+                content: vec![ContentItem::InputText { text: ui }],
+            });
+        }
+        input_with_instructions.extend(self.input.clone());
+        input_with_instructions
+    }
 }
 
 #[derive(Debug)]
@@ -71,6 +143,7 @@ pub enum ResponseEvent {
     },
     OutputTextDelta(String),
     ReasoningSummaryDelta(String),
+    ReasoningContentDelta(String),
 }
 
 #[derive(Debug, Serialize)]
@@ -144,14 +217,12 @@ pub(crate) struct ResponsesApiRequest<'a> {
     pub(crate) include: Vec<String>,
 }
 
-use crate::config::Config;
-
 pub(crate) fn create_reasoning_param_for_request(
-    config: &Config,
+    model_family: &ModelFamily,
     effort: ReasoningEffortConfig,
     summary: ReasoningSummaryConfig,
 ) -> Option<Reasoning> {
-    if model_supports_reasoning_summaries(config) {
+    if model_family.supports_reasoning_summaries {
         let effort: Option<OpenAiReasoningEffort> = effort.into();
         let effort = effort?;
         Some(Reasoning {
@@ -163,27 +234,6 @@ pub(crate) fn create_reasoning_param_for_request(
     }
 }
 
-pub fn model_supports_reasoning_summaries(config: &Config) -> bool {
-    // Currently, we hardcode this rule to decide whether to enable reasoning.
-    // We expect reasoning to apply only to OpenAI models, but we do not want
-    // users to have to mess with their config to disable reasoning for models
-    // that do not support it, such as `gpt-4.1`.
-    //
-    // Though if a user is using Codex with non-OpenAI models that, say, happen
-    // to start with "o", then they can set `model_reasoning_effort = "none"` in
-    // config.toml to disable reasoning.
-    //
-    // Converseley, if a user has a non-OpenAI provider that supports reasoning,
-    // they can set the top-level `model_supports_reasoning_summaries = true`
-    // config option to enable reasoning.
-    if config.model_supports_reasoning_summaries {
-        return true;
-    }
-
-    let model = &config.model;
-    model.starts_with("o") || model.starts_with("codex")
-}
-
 pub(crate) struct ResponseStream {
     pub(crate) rx_event: mpsc::Receiver<Result<ResponseEvent>>,
 }
@@ -198,6 +248,9 @@ impl Stream for ResponseStream {
 
 #[cfg(test)]
 mod tests {
+    #![allow(clippy::expect_used)]
+    use crate::model_family::find_family_for_model;
+
     use super::*;
 
     #[test]
@@ -207,7 +260,8 @@ mod tests {
             ..Default::default()
         };
         let expected = format!("{BASE_INSTRUCTIONS}\n{APPLY_PATCH_TOOL_INSTRUCTIONS}");
-        let full = prompt.get_full_instructions("gpt-4.1");
+        let model_family = find_family_for_model("gpt-4.1").expect("known model slug");
+        let full = prompt.get_full_instructions(&model_family);
         assert_eq!(full, expected);
     }
 }

codex-rs/core/src/codex.rs

@@ -37,6 +37,7 @@ use crate::apply_patch::convert_apply_patch_to_protocol;
 use crate::apply_patch::get_writable_roots;
 use crate::apply_patch::{self};
 use crate::client::ModelClient;
+use crate::client_common::EnvironmentContext;
 use crate::client_common::Prompt;
 use crate::client_common::ResponseEvent;
 use crate::config::Config;
@@ -56,16 +57,21 @@ use crate::mcp_tool_call::handle_mcp_tool_call;
 use crate::models::ContentItem;
 use crate::models::FunctionCallOutputPayload;
 use crate::models::LocalShellAction;
+use crate::models::ReasoningItemContent;
 use crate::models::ReasoningItemReasoningSummary;
 use crate::models::ResponseInputItem;
 use crate::models::ResponseItem;
 use crate::models::ShellToolCallParams;
+use crate::openai_tools::ToolsConfig;
+use crate::openai_tools::get_openai_tools;
 use crate::plan_tool::handle_update_plan;
 use crate::project_doc::get_user_instructions;
 use crate::protocol::AgentMessageDeltaEvent;
 use crate::protocol::AgentMessageEvent;
 use crate::protocol::AgentReasoningDeltaEvent;
 use crate::protocol::AgentReasoningEvent;
+use crate::protocol::AgentReasoningRawContentDeltaEvent;
+use crate::protocol::AgentReasoningRawContentEvent;
 use crate::protocol::ApplyPatchApprovalRequestEvent;
 use crate::protocol::AskForApproval;
 use crate::protocol::BackgroundEventEvent;
@@ -213,6 +219,7 @@ pub(crate) struct Session {
     shell_environment_policy: ShellEnvironmentPolicy,
     pub(crate) writable_roots: Mutex<Vec<PathBuf>>,
     disable_response_storage: bool,
+    tools_config: ToolsConfig,
 
     /// Manager for external MCP servers/tools.
     mcp_connection_manager: McpConnectionManager,
@@ -227,6 +234,7 @@ pub(crate) struct Session {
     state: Mutex<State>,
     codex_linux_sandbox_exe: Option<PathBuf>,
     user_shell: shell::Shell,
+    show_raw_agent_reasoning: bool,
 }
 
 impl Session {
@@ -806,6 +814,12 @@ async fn submission_loop(
                 let default_shell = shell::default_user_shell().await;
                 sess = Some(Arc::new(Session {
                     client,
+                    tools_config: ToolsConfig::new(
+                        &config.model_family,
+                        approval_policy,
+                        sandbox_policy.clone(),
+                        config.include_plan_tool,
+                    ),
                     tx_event: tx_event.clone(),
                     ctrl_c: Arc::clone(&ctrl_c),
                     user_instructions,
@@ -822,6 +836,7 @@ async fn submission_loop(
                     codex_linux_sandbox_exe: config.codex_linux_sandbox_exe.clone(),
                     disable_response_storage,
                     user_shell: default_shell,
+                    show_raw_agent_reasoning: config.show_raw_agent_reasoning,
                 }));
 
                 // Patch restored state into the newly created session.
@@ -1132,6 +1147,7 @@ async fn run_task(sess: Arc<Session>, sub_id: String, input: Vec<InputItem>) {
                             ResponseItem::Reasoning {
                                 id,
                                 summary,
+                                content,
                                 encrypted_content,
                             },
                             None,
@@ -1139,6 +1155,7 @@ async fn run_task(sess: Arc<Session>, sub_id: String, input: Vec<InputItem>) {
                             items_to_record_in_conversation_history.push(ResponseItem::Reasoning {
                                 id: id.clone(),
                                 summary: summary.clone(),
+                                content: content.clone(),
                                 encrypted_content: encrypted_content.clone(),
                             });
                         }
@@ -1197,13 +1214,22 @@ async fn run_turn(
     sub_id: String,
     input: Vec<ResponseItem>,
 ) -> CodexResult<Vec<ProcessedResponseItem>> {
-    let extra_tools = sess.mcp_connection_manager.list_all_tools();
+    let tools = get_openai_tools(
+        &sess.tools_config,
+        Some(sess.mcp_connection_manager.list_all_tools()),
+    );
+
     let prompt = Prompt {
         input,
         user_instructions: sess.user_instructions.clone(),
         store: !sess.disable_response_storage,
-        extra_tools,
+        tools,
         base_instructions_override: sess.base_instructions.clone(),
+        environment_context: Some(EnvironmentContext {
+            cwd: sess.cwd.clone(),
+            approval_policy: sess.approval_policy,
+            sandbox_policy: sess.sandbox_policy.clone(),
+        }),
     };
 
     let mut retries = 0;
@@ -1392,6 +1418,17 @@ async fn try_run_turn(
                 };
                 sess.tx_event.send(event).await.ok();
             }
+            ResponseEvent::ReasoningContentDelta(delta) => {
+                if sess.show_raw_agent_reasoning {
+                    let event = Event {
+                        id: sub_id.to_string(),
+                        msg: EventMsg::AgentReasoningRawContentDelta(
+                            AgentReasoningRawContentDeltaEvent { delta },
+                        ),
+                    };
+                    sess.tx_event.send(event).await.ok();
+                }
+            }
         }
     }
 }
@@ -1418,7 +1455,8 @@ async fn run_compact_task(
         input: turn_input,
         user_instructions: None,
         store: !sess.disable_response_storage,
-        extra_tools: HashMap::new(),
+        environment_context: None,
+        tools: Vec::new(),
         base_instructions_override: Some(compact_instructions.clone()),
     };
 
@@ -1498,7 +1536,12 @@ async fn handle_response_item(
             }
             None
         }
-        ResponseItem::Reasoning { summary, .. } => {
+        ResponseItem::Reasoning {
+            id: _,
+            summary,
+            content,
+            encrypted_content: _,
+        } => {
             for item in summary {
                 let text = match item {
                     ReasoningItemReasoningSummary::SummaryText { text } => text,
@@ -1509,6 +1552,21 @@ async fn handle_response_item(
                 };
                 sess.tx_event.send(event).await.ok();
             }
+            if sess.show_raw_agent_reasoning && content.is_some() {
+                let content = content.unwrap();
+                for item in content {
+                    let text = match item {
+                        ReasoningItemContent::ReasoningText { text } => text,
+                    };
+                    let event = Event {
+                        id: sub_id.to_string(),
+                        msg: EventMsg::AgentReasoningRawContent(AgentReasoningRawContentEvent {
+                            text,
+                        }),
+                    };
+                    sess.tx_event.send(event).await.ok();
+                }
+            }
             None
         }
         ResponseItem::FunctionCall {
@@ -1542,6 +1600,8 @@ async fn handle_response_item(
                 command: action.command,
                 workdir: action.working_directory,
                 timeout_ms: action.timeout_ms,
+                with_escalated_permissions: None,
+                justification: None,
             };
             let effective_call_id = match (call_id, id) {
                 (Some(call_id), _) => call_id,
@@ -1630,6 +1690,8 @@ fn to_exec_params(params: ShellToolCallParams, sess: &Session) -> ExecParams {
         cwd: sess.resolve_path(params.workdir.clone()),
         timeout_ms: params.timeout_ms,
         env: create_env(&sess.shell_environment_policy),
+        with_escalated_permissions: params.with_escalated_permissions,
+        justification: params.justification,
     }
 }
 
@@ -1730,13 +1792,19 @@ async fn handle_container_exec_with_params(
                 cwd: cwd.clone(),
                 timeout_ms: params.timeout_ms,
                 env: HashMap::new(),
+                with_escalated_permissions: params.with_escalated_permissions,
+                justification: params.justification.clone(),
             };
             let safety = if *user_explicitly_approved_this_action {
                 SafetyCheck::AutoApprove {
                     sandbox_type: SandboxType::None,
                 }
             } else {
-                assess_safety_for_untrusted_command(sess.approval_policy, &sess.sandbox_policy)
+                assess_safety_for_untrusted_command(
+                    sess.approval_policy,
+                    &sess.sandbox_policy,
+                    params.with_escalated_permissions.unwrap_or(false),
+                )
             };
             (
                 params,
@@ -1752,6 +1820,7 @@ async fn handle_container_exec_with_params(
                     sess.approval_policy,
                     &sess.sandbox_policy,
                     &state.approved_commands,
+                    params.with_escalated_permissions.unwrap_or(false),
                 )
             };
             let command_for_display = params.command.clone();
@@ -1768,7 +1837,7 @@ async fn handle_container_exec_with_params(
                     call_id.clone(),
                     params.command.clone(),
                     params.cwd.clone(),
-                    None,
+                    params.justification.clone(),
                 )
                 .await;
             match rx_approve.await.unwrap_or_default() {
@@ -1906,13 +1975,31 @@ async fn handle_sandbox_error(
     let cwd = exec_command_context.cwd.clone();
     let is_apply_patch = exec_command_context.apply_patch.is_some();
 
-    // Early out if the user never wants to be asked for approval; just return to the model immediately
-    if sess.approval_policy == AskForApproval::Never {
+    // Early out if either the user never wants to be asked for approval, or
+    // we're letting the model manage escalation requests. Otherwise, continue
+    match sess.approval_policy {
+        AskForApproval::Never | AskForApproval::OnRequest => {
+            return ResponseInputItem::FunctionCallOutput {
+                call_id,
+                output: FunctionCallOutputPayload {
+                    content: format!(
+                        "failed in sandbox {sandbox_type:?} with execution error: {error}"
+                    ),
+                    success: Some(false),
+                },
+            };
+        }
+        AskForApproval::UnlessTrusted | AskForApproval::OnFailure => (),
+    }
+
+    // similarly, if the command timed out, we can simply return this failure to the model
+    if matches!(error, SandboxErr::Timeout) {
         return ResponseInputItem::FunctionCallOutput {
             call_id,
             output: FunctionCallOutputPayload {
                 content: format!(
-                    "failed in sandbox {sandbox_type:?} with execution error: {error}"
+                    "command timed out after {} milliseconds",
+                    params.timeout_duration().as_millis()
                 ),
                 success: Some(false),
             },

codex-rs/core/src/config.rs

@@ -10,6 +10,8 @@ use crate::config_types::ShellEnvironmentPolicyToml;
 use crate::config_types::Tui;
 use crate::config_types::UriBasedFileOpener;
 use crate::flags::OPENAI_DEFAULT_MODEL;
+use crate::model_family::ModelFamily;
+use crate::model_family::find_family_for_model;
 use crate::model_provider_info::ModelProviderInfo;
 use crate::model_provider_info::built_in_model_providers;
 use crate::openai_model_info::get_model_info;
@@ -33,6 +35,8 @@ pub struct Config {
     /// Optional override of model selection.
     pub model: String,
 
+    pub model_family: ModelFamily,
+
     /// Size of the context window for the model, in tokens.
     pub model_context_window: Option<u64>,
 
@@ -57,12 +61,16 @@ pub struct Config {
     /// users are only interested in the final agent responses.
     pub hide_agent_reasoning: bool,
 
+    /// When set to `true`, `AgentReasoningRawContentEvent` events will be shown in the UI/output.
+    /// Defaults to `false`.
+    pub show_raw_agent_reasoning: bool,
+
     /// Disable server-side response storage (sends the full conversation
     /// context with every request). Currently necessary for OpenAI customers
     /// who have opted into Zero Data Retention (ZDR).
     pub disable_response_storage: bool,
 
-    /// User-provided instructions from instructions.md.
+    /// User-provided instructions from AGENTS.md.
     pub user_instructions: Option<String>,
 
     /// Base instructions override.
@@ -134,10 +142,6 @@ pub struct Config {
     /// request using the Responses API.
     pub model_reasoning_summary: ReasoningSummary,
 
-    /// When set to `true`, overrides the default heuristic and forces
-    /// `model_supports_reasoning_summaries()` to return `true`.
-    pub model_supports_reasoning_summaries: bool,
-
     /// Base URL for requests to ChatGPT (as opposed to the OpenAI API).
     pub chatgpt_base_url: String,
 
@@ -325,6 +329,10 @@ pub struct ConfigToml {
     /// UI/output. Defaults to `false`.
     pub hide_agent_reasoning: Option<bool>,
 
+    /// When set to `true`, `AgentReasoningRawContentEvent` events will be shown in the UI/output.
+    /// Defaults to `false`.
+    pub show_raw_agent_reasoning: Option<bool>,
+
     pub model_reasoning_effort: Option<ReasoningEffort>,
     pub model_reasoning_summary: Option<ReasoningSummary>,
 
@@ -377,6 +385,8 @@ pub struct ConfigOverrides {
     pub codex_linux_sandbox_exe: Option<PathBuf>,
     pub base_instructions: Option<String>,
     pub include_plan_tool: Option<bool>,
+    pub disable_response_storage: Option<bool>,
+    pub show_raw_agent_reasoning: Option<bool>,
 }
 
 impl Config {
@@ -400,6 +410,8 @@ impl Config {
             codex_linux_sandbox_exe,
             base_instructions,
             include_plan_tool,
+            disable_response_storage,
+            show_raw_agent_reasoning,
         } = overrides;
 
         let config_profile = match config_profile_key.as_ref().or(cfg.profile.as_ref()) {
@@ -465,7 +477,19 @@ impl Config {
             .or(config_profile.model)
             .or(cfg.model)
             .unwrap_or_else(default_model);
-        let openai_model_info = get_model_info(&model);
+        let model_family = find_family_for_model(&model).unwrap_or_else(|| {
+            let supports_reasoning_summaries =
+                cfg.model_supports_reasoning_summaries.unwrap_or(false);
+            ModelFamily {
+                slug: model.clone(),
+                family: model.clone(),
+                needs_special_apply_patch_instructions: false,
+                supports_reasoning_summaries,
+                uses_local_shell_tool: false,
+            }
+        });
+
+        let openai_model_info = get_model_info(&model_family);
         let model_context_window = cfg
             .model_context_window
             .or_else(|| openai_model_info.as_ref().map(|info| info.context_window));
@@ -490,6 +514,7 @@ impl Config {
 
         let config = Self {
             model,
+            model_family,
             model_context_window,
             model_max_output_tokens,
             model_provider_id,
@@ -504,6 +529,7 @@ impl Config {
             disable_response_storage: config_profile
                 .disable_response_storage
                 .or(cfg.disable_response_storage)
+                .or(disable_response_storage)
                 .unwrap_or(false),
             notify: cfg.notify,
             user_instructions,
@@ -518,6 +544,10 @@ impl Config {
             codex_linux_sandbox_exe,
 
             hide_agent_reasoning: cfg.hide_agent_reasoning.unwrap_or(false),
+            show_raw_agent_reasoning: cfg
+                .show_raw_agent_reasoning
+                .or(show_raw_agent_reasoning)
+                .unwrap_or(false),
             model_reasoning_effort: config_profile
                 .model_reasoning_effort
                 .or(cfg.model_reasoning_effort)
@@ -527,10 +557,6 @@ impl Config {
                 .or(cfg.model_reasoning_summary)
                 .unwrap_or_default(),
 
-            model_supports_reasoning_summaries: cfg
-                .model_supports_reasoning_summaries
-                .unwrap_or(false),
-
             chatgpt_base_url: config_profile
                 .chatgpt_base_url
                 .or(cfg.chatgpt_base_url)
@@ -549,7 +575,7 @@ impl Config {
             None => return None,
         };
 
-        p.push("instructions.md");
+        p.push("AGENTS.md");
         std::fs::read_to_string(&p).ok().and_then(|s| {
             let s = s.trim();
             if s.is_empty() {
@@ -816,7 +842,7 @@ disable_response_storage = true
             request_max_retries: Some(4),
             stream_max_retries: Some(10),
             stream_idle_timeout_ms: Some(300_000),
-            requires_auth: false,
+            requires_openai_auth: false,
         };
         let model_provider_map = {
             let mut model_provider_map = built_in_model_providers();
@@ -871,6 +897,7 @@ disable_response_storage = true
         assert_eq!(
             Config {
                 model: "o3".to_string(),
+                model_family: find_family_for_model("o3").expect("known model slug"),
                 model_context_window: Some(200_000),
                 model_max_output_tokens: Some(100_000),
                 model_provider_id: "openai".to_string(),
@@ -891,9 +918,9 @@ disable_response_storage = true
                 tui: Tui::default(),
                 codex_linux_sandbox_exe: None,
                 hide_agent_reasoning: false,
+                show_raw_agent_reasoning: false,
                 model_reasoning_effort: ReasoningEffort::High,
                 model_reasoning_summary: ReasoningSummary::Detailed,
-                model_supports_reasoning_summaries: false,
                 chatgpt_base_url: "https://chatgpt.com/backend-api/".to_string(),
                 experimental_resume: None,
                 base_instructions: None,
@@ -921,6 +948,7 @@ disable_response_storage = true
         )?;
         let expected_gpt3_profile_config = Config {
             model: "gpt-3.5-turbo".to_string(),
+            model_family: find_family_for_model("gpt-3.5-turbo").expect("known model slug"),
             model_context_window: Some(16_385),
             model_max_output_tokens: Some(4_096),
             model_provider_id: "openai-chat-completions".to_string(),
@@ -941,9 +969,9 @@ disable_response_storage = true
             tui: Tui::default(),
             codex_linux_sandbox_exe: None,
             hide_agent_reasoning: false,
+            show_raw_agent_reasoning: false,
             model_reasoning_effort: ReasoningEffort::default(),
             model_reasoning_summary: ReasoningSummary::default(),
-            model_supports_reasoning_summaries: false,
             chatgpt_base_url: "https://chatgpt.com/backend-api/".to_string(),
             experimental_resume: None,
             base_instructions: None,
@@ -986,6 +1014,7 @@ disable_response_storage = true
         )?;
         let expected_zdr_profile_config = Config {
             model: "o3".to_string(),
+            model_family: find_family_for_model("o3").expect("known model slug"),
             model_context_window: Some(200_000),
             model_max_output_tokens: Some(100_000),
             model_provider_id: "openai".to_string(),
@@ -1006,9 +1035,9 @@ disable_response_storage = true
             tui: Tui::default(),
             codex_linux_sandbox_exe: None,
             hide_agent_reasoning: false,
+            show_raw_agent_reasoning: false,
             model_reasoning_effort: ReasoningEffort::default(),
             model_reasoning_summary: ReasoningSummary::default(),
-            model_supports_reasoning_summaries: false,
             chatgpt_base_url: "https://chatgpt.com/backend-api/".to_string(),
             experimental_resume: None,
             base_instructions: None,

codex-rs/core/src/exec.rs

@@ -49,6 +49,14 @@ pub struct ExecParams {
     pub cwd: PathBuf,
     pub timeout_ms: Option<u64>,
     pub env: HashMap<String, String>,
+    pub with_escalated_permissions: Option<bool>,
+    pub justification: Option<String>,
+}
+
+impl ExecParams {
+    pub fn timeout_duration(&self) -> Duration {
+        Duration::from_millis(self.timeout_ms.unwrap_or(DEFAULT_TIMEOUT_MS))
+    }
 }
 
 #[derive(Clone, Copy, Debug, PartialEq)]
@@ -83,11 +91,9 @@ pub async fn process_exec_tool_call(
     {
         SandboxType::None => exec(params, sandbox_policy, ctrl_c, stdout_stream.clone()).await,
         SandboxType::MacosSeatbelt => {
+            let timeout = params.timeout_duration();
             let ExecParams {
-                command,
-                cwd,
-                timeout_ms,
-                env,
+                command, cwd, env, ..
             } = params;
             let child = spawn_command_under_seatbelt(
                 command,
@@ -97,14 +103,12 @@ pub async fn process_exec_tool_call(
                 env,
             )
             .await?;
-            consume_truncated_output(child, ctrl_c, timeout_ms, stdout_stream.clone()).await
+            consume_truncated_output(child, ctrl_c, timeout, stdout_stream.clone()).await
         }
         SandboxType::LinuxSeccomp => {
+            let timeout = params.timeout_duration();
             let ExecParams {
-                command,
-                cwd,
-                timeout_ms,
-                env,
+                command, cwd, env, ..
             } = params;
 
             let codex_linux_sandbox_exe = codex_linux_sandbox_exe
@@ -120,7 +124,7 @@ pub async fn process_exec_tool_call(
             )
             .await?;
 
-            consume_truncated_output(child, ctrl_c, timeout_ms, stdout_stream).await
+            consume_truncated_output(child, ctrl_c, timeout, stdout_stream).await
         }
     };
     let duration = start.elapsed();
@@ -255,16 +259,16 @@ pub struct ExecToolCallOutput {
 }
 
 async fn exec(
-    ExecParams {
-        command,
-        cwd,
-        timeout_ms,
-        env,
-    }: ExecParams,
+    params: ExecParams,
     sandbox_policy: &SandboxPolicy,
     ctrl_c: Arc<Notify>,
     stdout_stream: Option<StdoutStream>,
 ) -> Result<RawExecToolCallOutput> {
+    let timeout = params.timeout_duration();
+    let ExecParams {
+        command, cwd, env, ..
+    } = params;
+
     let (program, args) = command.split_first().ok_or_else(|| {
         CodexErr::Io(io::Error::new(
             io::ErrorKind::InvalidInput,
@@ -282,7 +286,7 @@ async fn exec(
         env,
     )
     .await?;
-    consume_truncated_output(child, ctrl_c, timeout_ms, stdout_stream).await
+    consume_truncated_output(child, ctrl_c, timeout, stdout_stream).await
 }
 
 /// Consumes the output of a child process, truncating it so it is suitable for
@@ -290,7 +294,7 @@ async fn exec(
 pub(crate) async fn consume_truncated_output(
     mut child: Child,
     ctrl_c: Arc<Notify>,
-    timeout_ms: Option<u64>,
+    timeout: Duration,
     stdout_stream: Option<StdoutStream>,
 ) -> Result<RawExecToolCallOutput> {
     // Both stdout and stderr were configured with `Stdio::piped()`
@@ -324,7 +328,6 @@ pub(crate) async fn consume_truncated_output(
     ));
 
     let interrupted = ctrl_c.notified();
-    let timeout = Duration::from_millis(timeout_ms.unwrap_or(DEFAULT_TIMEOUT_MS));
     let exit_status = tokio::select! {
         result = tokio::time::timeout(timeout, child.wait()) => {
             match result {

codex-rs/core/src/git_info.rs

@@ -9,7 +9,7 @@ use tokio::time::timeout;
 /// Timeout for git commands to prevent freezing on large repositories
 const GIT_COMMAND_TIMEOUT: TokioDuration = TokioDuration::from_secs(5);
 
-#[derive(Serialize, Deserialize, Clone)]
+#[derive(Serialize, Deserialize, Clone, Debug)]
 pub struct GitInfo {
     /// Current commit hash (SHA)
     #[serde(skip_serializing_if = "Option::is_none")]

codex-rs/core/src/lib.rs

@@ -28,9 +28,12 @@ mod mcp_connection_manager;
 mod mcp_tool_call;
 mod message_history;
 mod model_provider_info;
+pub use model_provider_info::BUILT_IN_OSS_MODEL_PROVIDER_ID;
 pub use model_provider_info::ModelProviderInfo;
 pub use model_provider_info::WireApi;
 pub use model_provider_info::built_in_model_providers;
+pub use model_provider_info::create_oss_provider_with_base_url;
+pub mod model_family;
 mod models;
 mod openai_model_info;
 mod openai_tools;
@@ -45,7 +48,5 @@ pub mod spawn;
 pub mod turn_diff_tracker;
 mod user_notification;
 pub mod util;
-
 pub use apply_patch::CODEX_APPLY_PATCH_ARG1;
-pub use client_common::model_supports_reasoning_summaries;
 pub use safety::get_platform_sandbox;

codex-rs/core/src/model_family.rs

@@ -0,0 +1,100 @@
+/// A model family is a group of models that share certain characteristics.
+#[derive(Debug, Clone, PartialEq, Eq, Hash)]
+pub struct ModelFamily {
+    /// The full model slug used to derive this model family, e.g.
+    /// "gpt-4.1-2025-04-14".
+    pub slug: String,
+
+    /// The model family name, e.g. "gpt-4.1". Note this should able to be used
+    /// with [`crate::openai_model_info::get_model_info`].
+    pub family: String,
+
+    /// True if the model needs additional instructions on how to use the
+    /// "virtual" `apply_patch` CLI.
+    pub needs_special_apply_patch_instructions: bool,
+
+    // Whether the `reasoning` field can be set when making a request to this
+    // model family. Note it has `effort` and `summary` subfields (though
+    // `summary` is optional).
+    pub supports_reasoning_summaries: bool,
+
+    // This should be set to true when the model expects a tool named
+    // "local_shell" to be provided. Its contract must be understood natively by
+    // the model such that its description can be omitted.
+    // See https://platform.openai.com/docs/guides/tools-local-shell
+    pub uses_local_shell_tool: bool,
+}
+
+macro_rules! model_family {
+    (
+        $slug:expr, $family:expr $(, $key:ident : $value:expr )* $(,)?
+    ) => {{
+        // defaults
+        let mut mf = ModelFamily {
+            slug: $slug.to_string(),
+            family: $family.to_string(),
+            needs_special_apply_patch_instructions: false,
+            supports_reasoning_summaries: false,
+            uses_local_shell_tool: false,
+        };
+        // apply overrides
+        $(
+            mf.$key = $value;
+        )*
+        Some(mf)
+    }};
+}
+
+macro_rules! simple_model_family {
+    (
+        $slug:expr, $family:expr
+    ) => {{
+        Some(ModelFamily {
+            slug: $slug.to_string(),
+            family: $family.to_string(),
+            needs_special_apply_patch_instructions: false,
+            supports_reasoning_summaries: false,
+            uses_local_shell_tool: false,
+        })
+    }};
+}
+
+/// Returns a `ModelFamily` for the given model slug, or `None` if the slug
+/// does not match any known model family.
+pub fn find_family_for_model(slug: &str) -> Option<ModelFamily> {
+    if slug.starts_with("o3") {
+        model_family!(
+            slug, "o3",
+            supports_reasoning_summaries: true,
+        )
+    } else if slug.starts_with("o4-mini") {
+        model_family!(
+            slug, "o4-mini",
+            supports_reasoning_summaries: true,
+        )
+    } else if slug.starts_with("codex-mini-latest") {
+        model_family!(
+            slug, "codex-mini-latest",
+            supports_reasoning_summaries: true,
+            uses_local_shell_tool: true,
+        )
+    } else if slug.starts_with("gpt-4.1") {
+        model_family!(
+            slug, "gpt-4.1",
+            needs_special_apply_patch_instructions: true,
+        )
+    } else if slug.starts_with("gpt-4o") {
+        simple_model_family!(slug, "gpt-4o")
+    } else if slug.starts_with("gpt-oss") {
+        simple_model_family!(slug, "gpt-oss")
+    } else if slug.starts_with("gpt-3.5") {
+        simple_model_family!(slug, "gpt-3.5")
+    } else if slug.starts_with("2025-08-06-model") {
+        model_family!(
+            slug, "2025-08-06-model",
+            supports_reasoning_summaries: true,
+        )
+    } else {
+        None
+    }
+}

codex-rs/core/src/model_provider_info.rs

@@ -9,7 +9,6 @@ use codex_login::AuthMode;
 use codex_login::CodexAuth;
 use serde::Deserialize;
 use serde::Serialize;
-use std::borrow::Cow;
 use std::collections::HashMap;
 use std::env::VarError;
 use std::time::Duration;
@@ -79,7 +78,7 @@ pub struct ModelProviderInfo {
 
     /// Whether this provider requires some form of standard authentication (API key, ChatGPT token).
     #[serde(default)]
-    pub requires_auth: bool,
+    pub requires_openai_auth: bool,
 }
 
 impl ModelProviderInfo {
@@ -87,26 +86,32 @@ impl ModelProviderInfo {
     /// reqwest Client applying:
     ///   • provider-specific headers (static + env based)
     ///   • Bearer auth header when an API key is available.
+    ///   • Auth token for OAuth.
     ///
-    /// When `require_api_key` is true and the provider declares an `env_key`
-    /// but the variable is missing/empty, returns an [`Err`] identical to the
+    /// If the provider declares an `env_key` but the variable is missing/empty, returns an [`Err`] identical to the
     /// one produced by [`ModelProviderInfo::api_key`].
     pub async fn create_request_builder<'a>(
         &'a self,
         client: &'a reqwest::Client,
         auth: &Option<CodexAuth>,
     ) -> crate::error::Result<reqwest::RequestBuilder> {
-        let auth: Cow<'_, Option<CodexAuth>> = if auth.is_some() {
-            Cow::Borrowed(auth)
-        } else {
-            Cow::Owned(self.get_fallback_auth()?)
+        let effective_auth = match self.api_key() {
+            Ok(Some(key)) => Some(CodexAuth::from_api_key(key)),
+            Ok(None) => auth.clone(),
+            Err(err) => {
+                if auth.is_some() {
+                    auth.clone()
+                } else {
+                    return Err(err);
+                }
+            }
         };
 
-        let url = self.get_full_url(&auth);
+        let url = self.get_full_url(&effective_auth);
 
         let mut builder = client.post(url);
 
-        if let Some(auth) = auth.as_ref() {
+        if let Some(auth) = effective_auth.as_ref() {
             builder = builder.bearer_auth(auth.get_token().await?);
         }
 
@@ -216,68 +221,105 @@ impl ModelProviderInfo {
             .map(Duration::from_millis)
             .unwrap_or(Duration::from_millis(DEFAULT_STREAM_IDLE_TIMEOUT_MS))
     }
-
-    fn get_fallback_auth(&self) -> crate::error::Result<Option<CodexAuth>> {
-        let api_key = self.api_key()?;
-        if let Some(api_key) = api_key {
-            return Ok(Some(CodexAuth::from_api_key(api_key)));
-        }
-        Ok(None)
-    }
 }
 
+const DEFAULT_OLLAMA_PORT: u32 = 11434;
+
+pub const BUILT_IN_OSS_MODEL_PROVIDER_ID: &str = "oss";
+
 /// Built-in default provider list.
 pub fn built_in_model_providers() -> HashMap<String, ModelProviderInfo> {
     use ModelProviderInfo as P;
 
     // We do not want to be in the business of adjucating which third-party
-    // providers are bundled with Codex CLI, so we only include the OpenAI
-    // provider by default. Users are encouraged to add to `model_providers`
-    // in config.toml to add their own providers.
-    [(
-        "openai",
-        P {
-            name: "OpenAI".into(),
-            // Allow users to override the default OpenAI endpoint by
-            // exporting `OPENAI_BASE_URL`. This is useful when pointing
-            // Codex at a proxy, mock server, or Azure-style deployment
-            // without requiring a full TOML override for the built-in
-            // OpenAI provider.
-            base_url: std::env::var("OPENAI_BASE_URL")
-                .ok()
-                .filter(|v| !v.trim().is_empty()),
-            env_key: None,
-            env_key_instructions: None,
-            wire_api: WireApi::Responses,
-            query_params: None,
-            http_headers: Some(
-                [("version".to_string(), env!("CARGO_PKG_VERSION").to_string())]
+    // providers are bundled with Codex CLI, so we only include the OpenAI and
+    // open source ("oss") providers by default. Users are encouraged to add to
+    // `model_providers` in config.toml to add their own providers.
+    [
+        (
+            "openai",
+            P {
+                name: "OpenAI".into(),
+                // Allow users to override the default OpenAI endpoint by
+                // exporting `OPENAI_BASE_URL`. This is useful when pointing
+                // Codex at a proxy, mock server, or Azure-style deployment
+                // without requiring a full TOML override for the built-in
+                // OpenAI provider.
+                base_url: std::env::var("OPENAI_BASE_URL")
+                    .ok()
+                    .filter(|v| !v.trim().is_empty()),
+                env_key: None,
+                env_key_instructions: None,
+                wire_api: WireApi::Responses,
+                query_params: None,
+                http_headers: Some(
+                    [("version".to_string(), env!("CARGO_PKG_VERSION").to_string())]
+                        .into_iter()
+                        .collect(),
+                ),
+                env_http_headers: Some(
+                    [
+                        (
+                            "OpenAI-Organization".to_string(),
+                            "OPENAI_ORGANIZATION".to_string(),
+                        ),
+                        ("OpenAI-Project".to_string(), "OPENAI_PROJECT".to_string()),
+                    ]
                     .into_iter()
                     .collect(),
-            ),
-            env_http_headers: Some(
-                [
-                    (
-                        "OpenAI-Organization".to_string(),
-                        "OPENAI_ORGANIZATION".to_string(),
-                    ),
-                    ("OpenAI-Project".to_string(), "OPENAI_PROJECT".to_string()),
-                ]
-                .into_iter()
-                .collect(),
-            ),
-            // Use global defaults for retry/timeout unless overridden in config.toml.
-            request_max_retries: None,
-            stream_max_retries: None,
-            stream_idle_timeout_ms: None,
-            requires_auth: true,
-        },
-    )]
+                ),
+                // Use global defaults for retry/timeout unless overridden in config.toml.
+                request_max_retries: None,
+                stream_max_retries: None,
+                stream_idle_timeout_ms: None,
+                requires_openai_auth: true,
+            },
+        ),
+        (BUILT_IN_OSS_MODEL_PROVIDER_ID, create_oss_provider()),
+    ]
     .into_iter()
     .map(|(k, v)| (k.to_string(), v))
     .collect()
 }
 
+pub fn create_oss_provider() -> ModelProviderInfo {
+    // These CODEX_OSS_ environment variables are experimental: we may
+    // switch to reading values from config.toml instead.
+    let codex_oss_base_url = match std::env::var("CODEX_OSS_BASE_URL")
+        .ok()
+        .filter(|v| !v.trim().is_empty())
+    {
+        Some(url) => url,
+        None => format!(
+            "http://localhost:{port}/v1",
+            port = std::env::var("CODEX_OSS_PORT")
+                .ok()
+                .filter(|v| !v.trim().is_empty())
+                .and_then(|v| v.parse::<u32>().ok())
+                .unwrap_or(DEFAULT_OLLAMA_PORT)
+        ),
+    };
+
+    create_oss_provider_with_base_url(&codex_oss_base_url)
+}
+
+pub fn create_oss_provider_with_base_url(base_url: &str) -> ModelProviderInfo {
+    ModelProviderInfo {
+        name: "gpt-oss".into(),
+        base_url: Some(base_url.into()),
+        env_key: None,
+        env_key_instructions: None,
+        wire_api: WireApi::Chat,
+        query_params: None,
+        http_headers: None,
+        env_http_headers: None,
+        request_max_retries: None,
+        stream_max_retries: None,
+        stream_idle_timeout_ms: None,
+        requires_openai_auth: false,
+    }
+}
+
 #[cfg(test)]
 mod tests {
     #![allow(clippy::unwrap_used)]
@@ -302,7 +344,7 @@ base_url = "http://localhost:11434/v1"
             request_max_retries: None,
             stream_max_retries: None,
             stream_idle_timeout_ms: None,
-            requires_auth: false,
+            requires_openai_auth: false,
         };
 
         let provider: ModelProviderInfo = toml::from_str(azure_provider_toml).unwrap();
@@ -331,7 +373,7 @@ query_params = { api-version = "2025-04-01-preview" }
             request_max_retries: None,
             stream_max_retries: None,
             stream_idle_timeout_ms: None,
-            requires_auth: false,
+            requires_openai_auth: false,
         };
 
         let provider: ModelProviderInfo = toml::from_str(azure_provider_toml).unwrap();
@@ -363,7 +405,7 @@ env_http_headers = { "X-Example-Env-Header" = "EXAMPLE_ENV_VAR" }
             request_max_retries: None,
             stream_max_retries: None,
             stream_idle_timeout_ms: None,
-            requires_auth: false,
+            requires_openai_auth: false,
         };
 
         let provider: ModelProviderInfo = toml::from_str(azure_provider_toml).unwrap();

codex-rs/core/src/models.rs

@@ -45,6 +45,8 @@ pub enum ResponseItem {
     Reasoning {
         id: String,
         summary: Vec<ReasoningItemReasoningSummary>,
+        #[serde(default, skip_serializing_if = "Option::is_none")]
+        content: Option<Vec<ReasoningItemContent>>,
         encrypted_content: Option<String>,
     },
     LocalShellCall {
@@ -136,6 +138,12 @@ pub enum ReasoningItemReasoningSummary {
     SummaryText { text: String },
 }
 
+#[derive(Debug, Clone, Serialize, Deserialize, PartialEq)]
+#[serde(tag = "type", rename_all = "snake_case")]
+pub enum ReasoningItemContent {
+    ReasoningText { text: String },
+}
+
 impl From<Vec<InputItem>> for ResponseInputItem {
     fn from(items: Vec<InputItem>) -> Self {
         Self::Message {
@@ -183,6 +191,10 @@ pub struct ShellToolCallParams {
     // The wire format uses `timeout`, which has ambiguous units, so we use
     // `timeout_ms` as the field name so it is clear in code.
     pub timeout_ms: Option<u64>,
+    #[serde(skip_serializing_if = "Option::is_none")]
+    pub with_escalated_permissions: Option<bool>,
+    #[serde(skip_serializing_if = "Option::is_none")]
+    pub justification: Option<String>,
 }
 
 #[derive(Debug, Clone, PartialEq)]
@@ -294,6 +306,8 @@ mod tests {
                 command: vec!["ls".to_string(), "-l".to_string()],
                 workdir: Some("/tmp".to_string()),
                 timeout_ms: Some(1000),
+                with_escalated_permissions: None,
+                justification: None,
             },
             params
         );

codex-rs/core/src/openai_model_info.rs

@@ -1,3 +1,5 @@
+use crate::model_family::ModelFamily;
+
 /// Metadata about a model, particularly OpenAI models.
 /// We may want to consider including details like the pricing for
 /// input tokens, output tokens, etc., though users will need to be able to
@@ -12,10 +14,19 @@ pub(crate) struct ModelInfo {
     pub(crate) max_output_tokens: u64,
 }
 
-/// Note details such as what a model like gpt-4o is aliased to may be out of
-/// date.
-pub(crate) fn get_model_info(name: &str) -> Option<ModelInfo> {
-    match name {
+pub(crate) fn get_model_info(model_family: &ModelFamily) -> Option<ModelInfo> {
+    match model_family.slug.as_str() {
+        // OSS models have a 128k shared token pool.
+        // Arbitrarily splitting it: 3/4 input context, 1/4 output.
+        // https://openai.com/index/gpt-oss-model-card/
+        "gpt-oss-20b" => Some(ModelInfo {
+            context_window: 96_000,
+            max_output_tokens: 32_000,
+        }),
+        "gpt-oss-120b" => Some(ModelInfo {
+            context_window: 96_000,
+            max_output_tokens: 32_000,
+        }),
         // https://platform.openai.com/docs/models/o3
         "o3" => Some(ModelInfo {
             context_window: 200_000,
@@ -66,6 +77,11 @@ pub(crate) fn get_model_info(name: &str) -> Option<ModelInfo> {
             max_output_tokens: 4_096,
         }),
 
+        "2025-08-06-model" => Some(ModelInfo {
+            context_window: 200_000,
+            max_output_tokens: 100_000,
+        }),
+
         _ => None,
     }
 }

codex-rs/core/src/openai_tools.rs

@@ -1,22 +1,28 @@
+use serde::Deserialize;
 use serde::Serialize;
 use serde_json::json;
 use std::collections::BTreeMap;
-use std::sync::LazyLock;
+use std::collections::HashMap;
 
-use crate::client_common::Prompt;
+use crate::model_family::ModelFamily;
 use crate::plan_tool::PLAN_TOOL;
+use crate::protocol::AskForApproval;
+use crate::protocol::SandboxPolicy;
 
-#[derive(Debug, Clone, Serialize)]
-pub(crate) struct ResponsesApiTool {
-    pub(crate) name: &'static str,
-    pub(crate) description: &'static str,
+#[derive(Debug, Clone, Serialize, PartialEq)]
+pub struct ResponsesApiTool {
+    pub(crate) name: String,
+    pub(crate) description: String,
+    /// TODO: Validation. When strict is set to true, the JSON schema,
+    /// `required` and `additional_properties` must be present. All fields in
+    /// `properties` must be present in `required`.
     pub(crate) strict: bool,
     pub(crate) parameters: JsonSchema,
 }
 
 /// When serialized as JSON, this produces a valid "Tool" in the OpenAI
 /// Responses API.
-#[derive(Debug, Clone, Serialize)]
+#[derive(Debug, Clone, Serialize, PartialEq)]
 #[serde(tag = "type")]
 pub(crate) enum OpenAiTool {
     #[serde(rename = "function")]
@@ -25,78 +31,217 @@ pub(crate) enum OpenAiTool {
     LocalShell {},
 }
 
+#[derive(Debug, Clone)]
+pub enum ConfigShellToolType {
+    DefaultShell,
+    ShellWithRequest { sandbox_policy: SandboxPolicy },
+    LocalShell,
+}
+
+#[derive(Debug, Clone)]
+pub struct ToolsConfig {
+    pub shell_type: ConfigShellToolType,
+    pub plan_tool: bool,
+}
+
+impl ToolsConfig {
+    pub fn new(
+        model_family: &ModelFamily,
+        approval_policy: AskForApproval,
+        sandbox_policy: SandboxPolicy,
+        include_plan_tool: bool,
+    ) -> Self {
+        let mut shell_type = if model_family.uses_local_shell_tool {
+            ConfigShellToolType::LocalShell
+        } else {
+            ConfigShellToolType::DefaultShell
+        };
+        if matches!(approval_policy, AskForApproval::OnRequest) {
+            shell_type = ConfigShellToolType::ShellWithRequest {
+                sandbox_policy: sandbox_policy.clone(),
+            }
+        }
+
+        Self {
+            shell_type,
+            plan_tool: include_plan_tool,
+        }
+    }
+}
+
 /// Generic JSON‑Schema subset needed for our tool definitions
-#[derive(Debug, Clone, Serialize)]
+#[derive(Debug, Clone, Serialize, Deserialize, PartialEq)]
 #[serde(tag = "type", rename_all = "lowercase")]
 pub(crate) enum JsonSchema {
-    String,
-    Number,
+    Boolean {
+        #[serde(skip_serializing_if = "Option::is_none")]
+        description: Option<String>,
+    },
+    String {
+        #[serde(skip_serializing_if = "Option::is_none")]
+        description: Option<String>,
+    },
+    Number {
+        #[serde(skip_serializing_if = "Option::is_none")]
+        description: Option<String>,
+    },
     Array {
         items: Box<JsonSchema>,
+
+        #[serde(skip_serializing_if = "Option::is_none")]
+        description: Option<String>,
     },
     Object {
         properties: BTreeMap<String, JsonSchema>,
-        required: &'static [&'static str],
-        #[serde(rename = "additionalProperties")]
-        additional_properties: bool,
+        #[serde(skip_serializing_if = "Option::is_none")]
+        required: Option<Vec<String>>,
+        #[serde(
+            rename = "additionalProperties",
+            skip_serializing_if = "Option::is_none"
+        )]
+        additional_properties: Option<bool>,
     },
 }
 
-/// Tool usage specification
-static DEFAULT_TOOLS: LazyLock<Vec<OpenAiTool>> = LazyLock::new(|| {
+fn create_shell_tool() -> OpenAiTool {
     let mut properties = BTreeMap::new();
     properties.insert(
         "command".to_string(),
         JsonSchema::Array {
-            items: Box::new(JsonSchema::String),
+            items: Box::new(JsonSchema::String { description: None }),
+            description: None,
         },
     );
-    properties.insert("workdir".to_string(), JsonSchema::String);
-    properties.insert("timeout".to_string(), JsonSchema::Number);
+    properties.insert(
+        "workdir".to_string(),
+        JsonSchema::String { description: None },
+    );
+    properties.insert(
+        "timeout".to_string(),
+        JsonSchema::Number { description: None },
+    );
 
-    vec![OpenAiTool::Function(ResponsesApiTool {
-        name: "shell",
-        description: "Runs a shell command, and returns its output.",
+    OpenAiTool::Function(ResponsesApiTool {
+        name: "shell".to_string(),
+        description: "Runs a shell command and returns its output".to_string(),
         strict: false,
         parameters: JsonSchema::Object {
             properties,
-            required: &["command"],
-            additional_properties: false,
+            required: Some(vec!["command".to_string()]),
+            additional_properties: Some(false),
         },
-    })]
-});
+    })
+}
 
-static DEFAULT_CODEX_MODEL_TOOLS: LazyLock<Vec<OpenAiTool>> =
-    LazyLock::new(|| vec![OpenAiTool::LocalShell {}]);
+fn create_shell_tool_for_sandbox(sandbox_policy: &SandboxPolicy) -> OpenAiTool {
+    let mut properties = BTreeMap::new();
+    properties.insert(
+        "command".to_string(),
+        JsonSchema::Array {
+            items: Box::new(JsonSchema::String { description: None }),
+            description: Some("The command to execute".to_string()),
+        },
+    );
+    properties.insert(
+        "workdir".to_string(),
+        JsonSchema::String {
+            description: Some("The working directory to execute the command in".to_string()),
+        },
+    );
+    properties.insert(
+        "timeout".to_string(),
+        JsonSchema::Number {
+            description: Some("The timeout for the command in milliseconds".to_string()),
+        },
+    );
+
+    if matches!(sandbox_policy, SandboxPolicy::WorkspaceWrite { .. }) {
+        properties.insert(
+        "with_escalated_permissions".to_string(),
+        JsonSchema::Boolean {
+            description: Some("Whether to request escalated permissions. Set to true if command needs to be run without sandbox restrictions".to_string()),
+        },
+    );
+        properties.insert(
+        "justification".to_string(),
+        JsonSchema::String {
+            description: Some("Only set if ask_for_escalated_permissions is true. 1-sentence explanation of why we want to run this command.".to_string()),
+        },
+    );
+    }
+
+    let description = match sandbox_policy {
+        SandboxPolicy::WorkspaceWrite {
+            network_access,
+            ..
+        } => {
+            format!(
+                r#"
+The shell tool is used to execute shell commands.
+- When invoking the shell tool, your call will be running in a landlock sandbox, and some shell commands will require escalated privileges:
+  - Types of actions that require escalated privileges:
+    - Reading files outside the current directory
+    - Writing files outside the current directory, and protected folders like .git or .env{}
+  - Examples of commands that require escalated privileges:
+    - git commit
+    - npm install or pnpm install
+    - cargo build
+    - cargo test
+- When invoking a command that will require escalated privileges:
+  - Provide the with_escalated_permissions parameter with the boolean value true
+  - Include a short, 1 sentence explanation for why we need to run with_escalated_permissions in the justification parameter."#,
+                if !network_access {
+                    "\n  - Commands that require network access\n"
+                } else {
+                    ""
+                }
+            )
+        }
+        SandboxPolicy::DangerFullAccess => {
+            "Runs a shell command and returns its output.".to_string()
+        }
+        SandboxPolicy::ReadOnly => {
+            r#"
+The shell tool is used to execute shell commands.
+- When invoking the shell tool, your call will be running in a landlock sandbox, and some shell commands (including apply_patch) will require escalated permissions:
+  - Types of actions that require escalated privileges:
+    - Reading files outside the current directory
+    - Writing files
+    - Applying patches
+  - Examples of commands that require escalated privileges:
+    - apply_patch
+    - git commit
+    - npm install or pnpm install
+    - cargo build
+    - cargo test
+- When invoking a command that will require escalated privileges:
+  - Provide the with_escalated_permissions parameter with the boolean value true
+  - Include a short, 1 sentence explanation for why we need to run with_escalated_permissions in the justification parameter"#.to_string()
+        }
+    };
+
+    OpenAiTool::Function(ResponsesApiTool {
+        name: "shell".to_string(),
+        description,
+        strict: false,
+        parameters: JsonSchema::Object {
+            properties,
+            required: Some(vec!["command".to_string()]),
+            additional_properties: Some(false),
+        },
+    })
+}
 
 /// Returns JSON values that are compatible with Function Calling in the
 /// Responses API:
 /// https://platform.openai.com/docs/guides/function-calling?api-mode=responses
 pub(crate) fn create_tools_json_for_responses_api(
-    prompt: &Prompt,
-    model: &str,
-    include_plan_tool: bool,
+    tools: &Vec<OpenAiTool>,
 ) -> crate::error::Result<Vec<serde_json::Value>> {
-    // Assemble tool list: built-in tools + any extra tools from the prompt.
-    let default_tools = if model.starts_with("codex") {
-        &DEFAULT_CODEX_MODEL_TOOLS
-    } else {
-        &DEFAULT_TOOLS
-    };
-    let mut tools_json = Vec::with_capacity(default_tools.len() + prompt.extra_tools.len());
-    for t in default_tools.iter() {
-        tools_json.push(serde_json::to_value(t)?);
-    }
-    tools_json.extend(
-        prompt
-            .extra_tools
-            .clone()
-            .into_iter()
-            .map(|(name, tool)| mcp_tool_to_openai_tool(name, tool)),
-    );
+    let mut tools_json = Vec::new();
 
-    if include_plan_tool {
-        tools_json.push(serde_json::to_value(PLAN_TOOL.clone())?);
+    for tool in tools {
+        tools_json.push(serde_json::to_value(tool)?);
     }
 
     Ok(tools_json)
@@ -106,14 +251,11 @@ pub(crate) fn create_tools_json_for_responses_api(
 /// Chat Completions API:
 /// https://platform.openai.com/docs/guides/function-calling?api-mode=chat
 pub(crate) fn create_tools_json_for_chat_completions_api(
-    prompt: &Prompt,
-    model: &str,
-    include_plan_tool: bool,
+    tools: &Vec<OpenAiTool>,
 ) -> crate::error::Result<Vec<serde_json::Value>> {
     // We start with the JSON for the Responses API and than rewrite it to match
     // the chat completions tool call format.
-    let responses_api_tools_json =
-        create_tools_json_for_responses_api(prompt, model, include_plan_tool)?;
+    let responses_api_tools_json = create_tools_json_for_responses_api(tools)?;
     let tools_json = responses_api_tools_json
         .into_iter()
         .filter_map(|mut tool| {
@@ -136,10 +278,10 @@ pub(crate) fn create_tools_json_for_chat_completions_api(
     Ok(tools_json)
 }
 
-fn mcp_tool_to_openai_tool(
+pub(crate) fn mcp_tool_to_openai_tool(
     fully_qualified_name: String,
     tool: mcp_types::Tool,
-) -> serde_json::Value {
+) -> Result<ResponsesApiTool, serde_json::Error> {
     let mcp_types::Tool {
         description,
         mut input_schema,
@@ -154,12 +296,205 @@ fn mcp_tool_to_openai_tool(
         input_schema.properties = Some(serde_json::Value::Object(serde_json::Map::new()));
     }
 
-    // TODO(mbolin): Change the contract of this function to return
-    // ResponsesApiTool.
-    json!({
-        "name": fully_qualified_name,
-        "description": description,
-        "parameters": input_schema,
-        "type": "function",
+    let serialized_input_schema = serde_json::to_value(input_schema)?;
+    let input_schema = serde_json::from_value::<JsonSchema>(serialized_input_schema)?;
+
+    Ok(ResponsesApiTool {
+        name: fully_qualified_name,
+        description: description.unwrap_or_default(),
+        strict: false,
+        parameters: input_schema,
     })
 }
+
+/// Returns a list of OpenAiTools based on the provided config and MCP tools.
+/// Note that the keys of mcp_tools should be fully qualified names. See
+/// [`McpConnectionManager`] for more details.
+pub(crate) fn get_openai_tools(
+    config: &ToolsConfig,
+    mcp_tools: Option<HashMap<String, mcp_types::Tool>>,
+) -> Vec<OpenAiTool> {
+    let mut tools: Vec<OpenAiTool> = Vec::new();
+
+    match &config.shell_type {
+        ConfigShellToolType::DefaultShell => {
+            tools.push(create_shell_tool());
+        }
+        ConfigShellToolType::ShellWithRequest { sandbox_policy } => {
+            tools.push(create_shell_tool_for_sandbox(sandbox_policy));
+        }
+        ConfigShellToolType::LocalShell => {
+            tools.push(OpenAiTool::LocalShell {});
+        }
+    }
+
+    if config.plan_tool {
+        tools.push(PLAN_TOOL.clone());
+    }
+
+    if let Some(mcp_tools) = mcp_tools {
+        for (name, tool) in mcp_tools {
+            match mcp_tool_to_openai_tool(name.clone(), tool.clone()) {
+                Ok(converted_tool) => tools.push(OpenAiTool::Function(converted_tool)),
+                Err(e) => {
+                    tracing::error!("Failed to convert {name:?} MCP tool to OpenAI tool: {e:?}");
+                }
+            }
+        }
+    }
+
+    tools
+}
+
+#[cfg(test)]
+#[allow(clippy::expect_used)]
+mod tests {
+    use crate::model_family::find_family_for_model;
+    use mcp_types::ToolInputSchema;
+
+    use super::*;
+
+    fn assert_eq_tool_names(tools: &[OpenAiTool], expected_names: &[&str]) {
+        let tool_names = tools
+            .iter()
+            .map(|tool| match tool {
+                OpenAiTool::Function(ResponsesApiTool { name, .. }) => name,
+                OpenAiTool::LocalShell {} => "local_shell",
+            })
+            .collect::<Vec<_>>();
+
+        assert_eq!(
+            tool_names.len(),
+            expected_names.len(),
+            "tool_name mismatch, {tool_names:?}, {expected_names:?}",
+        );
+        for (name, expected_name) in tool_names.iter().zip(expected_names.iter()) {
+            assert_eq!(
+                name, expected_name,
+                "tool_name mismatch, {name:?}, {expected_name:?}"
+            );
+        }
+    }
+
+    #[test]
+    fn test_get_openai_tools() {
+        let model_family = find_family_for_model("codex-mini-latest")
+            .expect("codex-mini-latest should be a valid model family");
+        let config = ToolsConfig::new(
+            &model_family,
+            AskForApproval::Never,
+            SandboxPolicy::ReadOnly,
+            true,
+        );
+        let tools = get_openai_tools(&config, Some(HashMap::new()));
+
+        assert_eq_tool_names(&tools, &["local_shell", "update_plan"]);
+    }
+
+    #[test]
+    fn test_get_openai_tools_default_shell() {
+        let model_family = find_family_for_model("o3").expect("o3 should be a valid model family");
+        let config = ToolsConfig::new(
+            &model_family,
+            AskForApproval::Never,
+            SandboxPolicy::ReadOnly,
+            true,
+        );
+        let tools = get_openai_tools(&config, Some(HashMap::new()));
+
+        assert_eq_tool_names(&tools, &["shell", "update_plan"]);
+    }
+
+    #[test]
+    fn test_get_openai_tools_mcp_tools() {
+        let model_family = find_family_for_model("o3").expect("o3 should be a valid model family");
+        let config = ToolsConfig::new(
+            &model_family,
+            AskForApproval::Never,
+            SandboxPolicy::ReadOnly,
+            false,
+        );
+        let tools = get_openai_tools(
+            &config,
+            Some(HashMap::from([(
+                "test_server/do_something_cool".to_string(),
+                mcp_types::Tool {
+                    name: "do_something_cool".to_string(),
+                    input_schema: ToolInputSchema {
+                        properties: Some(serde_json::json!({
+                            "string_argument": {
+                                "type": "string",
+                            },
+                            "number_argument": {
+                                "type": "number",
+                            },
+                            "object_argument": {
+                                "type": "object",
+                                "properties": {
+                                    "string_property": { "type": "string" },
+                                    "number_property": { "type": "number" },
+                                },
+                                "required": [
+                                    "string_property",
+                                    "number_property"
+                                ],
+                                "additionalProperties": Some(false),
+                            },
+                        })),
+                        required: None,
+                        r#type: "object".to_string(),
+                    },
+                    output_schema: None,
+                    title: None,
+                    annotations: None,
+                    description: Some("Do something cool".to_string()),
+                },
+            )])),
+        );
+
+        assert_eq_tool_names(&tools, &["shell", "test_server/do_something_cool"]);
+
+        assert_eq!(
+            tools[1],
+            OpenAiTool::Function(ResponsesApiTool {
+                name: "test_server/do_something_cool".to_string(),
+                parameters: JsonSchema::Object {
+                    properties: BTreeMap::from([
+                        (
+                            "string_argument".to_string(),
+                            JsonSchema::String { description: None }
+                        ),
+                        (
+                            "number_argument".to_string(),
+                            JsonSchema::Number { description: None }
+                        ),
+                        (
+                            "object_argument".to_string(),
+                            JsonSchema::Object {
+                                properties: BTreeMap::from([
+                                    (
+                                        "string_property".to_string(),
+                                        JsonSchema::String { description: None }
+                                    ),
+                                    (
+                                        "number_property".to_string(),
+                                        JsonSchema::Number { description: None }
+                                    ),
+                                ]),
+                                required: Some(vec![
+                                    "string_property".to_string(),
+                                    "number_property".to_string(),
+                                ]),
+                                additional_properties: Some(false),
+                            },
+                        ),
+                    ]),
+                    required: None,
+                    additional_properties: None,
+                },
+                description: "Do something cool".to_string(),
+                strict: false,
+            })
+        );
+    }
+}

codex-rs/core/src/plan_tool.rs

@@ -39,23 +39,30 @@ pub struct UpdatePlanArgs {
 
 pub(crate) static PLAN_TOOL: LazyLock<OpenAiTool> = LazyLock::new(|| {
     let mut plan_item_props = BTreeMap::new();
-    plan_item_props.insert("step".to_string(), JsonSchema::String);
-    plan_item_props.insert("status".to_string(), JsonSchema::String);
+    plan_item_props.insert("step".to_string(), JsonSchema::String { description: None });
+    plan_item_props.insert(
+        "status".to_string(),
+        JsonSchema::String { description: None },
+    );
 
     let plan_items_schema = JsonSchema::Array {
+        description: Some("The list of steps".to_string()),
         items: Box::new(JsonSchema::Object {
             properties: plan_item_props,
-            required: &["step", "status"],
-            additional_properties: false,
+            required: Some(vec!["step".to_string(), "status".to_string()]),
+            additional_properties: Some(false),
         }),
     };
 
     let mut properties = BTreeMap::new();
-    properties.insert("explanation".to_string(), JsonSchema::String);
+    properties.insert(
+        "explanation".to_string(),
+        JsonSchema::String { description: None },
+    );
     properties.insert("plan".to_string(), plan_items_schema);
 
     OpenAiTool::Function(ResponsesApiTool {
-        name: "update_plan",
+        name: "update_plan".to_string(),
         description: r#"Use the update_plan tool to keep the user updated on the current plan for the task.
 After understanding the user's task, call the update_plan tool with an initial plan. An example of a plan:
 1. Explore the codebase to find relevant files (status: in_progress)
@@ -66,12 +73,12 @@ Until all the steps are finished, there should always be exactly one in_progress
 Call the update_plan tool whenever you finish a step, marking the completed step as `completed` and marking the next step as `in_progress`.
 Before running a command, consider whether or not you have completed the previous step, and make sure to mark it as completed before moving on to the next step.
 Sometimes, you may need to change plans in the middle of a task: call `update_plan` with the updated plan and make sure to provide an `explanation` of the rationale when doing so.
-When all steps are completed, call update_plan one last time with all steps marked as `completed`."#,
+When all steps are completed, call update_plan one last time with all steps marked as `completed`."#.to_string(),
         strict: false,
         parameters: JsonSchema::Object {
             properties,
-            required: &["plan"],
-            additional_properties: false,
+            required: Some(vec!["plan".to_string()]),
+            additional_properties: Some(false),
         },
     })
 });

codex-rs/core/src/protocol.rs

@@ -150,13 +150,17 @@ pub enum AskForApproval {
     /// the user to approve execution without a sandbox.
     OnFailure,
 
+    /// The model decides when to ask the user for approval.
+    OnRequest,
+
     /// Never ask the user to approve commands. Failures are immediately returned
     /// to the model, and never escalated to the user for approval.
     Never,
 }
 
 /// Determines execution restrictions for model shell commands.
-#[derive(Debug, Clone, PartialEq, Eq, Serialize, Deserialize)]
+#[derive(Debug, Clone, PartialEq, Eq, Serialize, Deserialize, Display)]
+#[strum(serialize_all = "kebab-case")]
 #[serde(tag = "mode", rename_all = "kebab-case")]
 pub enum SandboxPolicy {
     /// No restrictions whatsoever. Use with caution.
@@ -359,6 +363,12 @@ pub enum EventMsg {
     /// Agent reasoning delta event from agent.
     AgentReasoningDelta(AgentReasoningDeltaEvent),
 
+    /// Raw chain-of-thought from agent.
+    AgentReasoningRawContent(AgentReasoningRawContentEvent),
+
+    /// Agent reasoning content delta event from agent.
+    AgentReasoningRawContentDelta(AgentReasoningRawContentDeltaEvent),
+
     /// Ack the client's configure message.
     SessionConfigured(SessionConfiguredEvent),
 
@@ -419,6 +429,12 @@ pub struct TokenUsage {
     pub total_tokens: u64,
 }
 
+impl TokenUsage {
+    pub fn is_zero(&self) -> bool {
+        self.total_tokens == 0
+    }
+}
+
 #[derive(Debug, Clone, Deserialize, Serialize)]
 pub struct FinalOutput {
     pub token_usage: TokenUsage,
@@ -464,6 +480,16 @@ pub struct AgentReasoningEvent {
     pub text: String,
 }
 
+#[derive(Debug, Clone, Deserialize, Serialize)]
+pub struct AgentReasoningRawContentEvent {
+    pub text: String,
+}
+
+#[derive(Debug, Clone, Deserialize, Serialize)]
+pub struct AgentReasoningRawContentDeltaEvent {
+    pub delta: String,
+}
+
 #[derive(Debug, Clone, Deserialize, Serialize)]
 pub struct AgentReasoningDeltaEvent {
     pub delta: String,
@@ -600,6 +626,41 @@ pub struct PatchApplyEndEvent {
     pub success: bool,
 }
 
+/// Kind of a single line in a diff hunk body.
+#[derive(Debug, Clone, Copy, Deserialize, Serialize, PartialEq, Eq)]
+#[serde(rename_all = "snake_case")]
+pub enum DiffLineKind {
+    /// Line was inserted (prefixed with '+').
+    Add,
+    /// Line was deleted (prefixed with '-').
+    Delete,
+    /// Context line (no prefix or a single space in unified diff output).
+    Context,
+}
+
+/// A single line within a diff hunk.
+#[derive(Debug, Clone, Deserialize, Serialize)]
+pub struct DiffLine {
+    pub kind: DiffLineKind,
+    /// Text content of the line without the unified diff prefix character.
+    pub text: String,
+}
+
+/// Structured representation of a unified diff hunk.
+#[derive(Debug, Clone, Deserialize, Serialize)]
+pub struct DiffHunk {
+    /// 1-based starting line number in the original file.
+    pub old_start: u32,
+    /// Number of lines in the original file that the hunk covers.
+    pub old_count: u32,
+    /// 1-based starting line number in the new file.
+    pub new_start: u32,
+    /// Number of lines in the new file that the hunk covers.
+    pub new_count: u32,
+    /// Lines within this hunk in unified diff order.
+    pub lines: Vec<DiffLine>,
+}
+
 #[derive(Debug, Clone, Deserialize, Serialize)]
 pub struct TurnDiffEvent {
     pub unified_diff: String,
@@ -659,8 +720,14 @@ pub enum FileChange {
     },
     Delete,
     Update {
+        /// Unified diff for this file change. Retained for compatibility with
+        /// older clients that do not support structured hunks.
         unified_diff: String,
         move_path: Option<PathBuf>,
+        /// Optional structured representation of the diff hunks to avoid
+        /// requiring clients to parse unified diff syntax.
+        #[serde(skip_serializing_if = "Option::is_none")]
+        hunks: Option<Vec<DiffHunk>>,
     },
 }
 

codex-rs/core/src/safety.rs

@@ -11,7 +11,7 @@ use crate::is_safe_command::is_known_safe_command;
 use crate::protocol::AskForApproval;
 use crate::protocol::SandboxPolicy;
 
-#[derive(Debug)]
+#[derive(Debug, PartialEq)]
 pub enum SafetyCheck {
     AutoApprove { sandbox_type: SandboxType },
     AskUser,
@@ -31,7 +31,7 @@ pub fn assess_patch_safety(
     }
 
     match policy {
-        AskForApproval::OnFailure | AskForApproval::Never => {
+        AskForApproval::OnFailure | AskForApproval::Never | AskForApproval::OnRequest => {
             // Continue to see if this can be auto-approved.
         }
         // TODO(ragona): I'm not sure this is actually correct? I believe in this case
@@ -76,6 +76,7 @@ pub fn assess_command_safety(
     approval_policy: AskForApproval,
     sandbox_policy: &SandboxPolicy,
     approved: &HashSet<Vec<String>>,
+    with_escalated_permissions: bool,
 ) -> SafetyCheck {
     // A command is "trusted" because either:
     // - it belongs to a set of commands we consider "safe" by default, or
@@ -96,12 +97,13 @@ pub fn assess_command_safety(
         };
     }
 
-    assess_safety_for_untrusted_command(approval_policy, sandbox_policy)
+    assess_safety_for_untrusted_command(approval_policy, sandbox_policy, with_escalated_permissions)
 }
 
 pub(crate) fn assess_safety_for_untrusted_command(
     approval_policy: AskForApproval,
     sandbox_policy: &SandboxPolicy,
+    with_escalated_permissions: bool,
 ) -> SafetyCheck {
     use AskForApproval::*;
     use SandboxPolicy::*;
@@ -113,9 +115,23 @@ pub(crate) fn assess_safety_for_untrusted_command(
             // commands.
             SafetyCheck::AskUser
         }
-        (OnFailure, DangerFullAccess) | (Never, DangerFullAccess) => SafetyCheck::AutoApprove {
+        (OnFailure, DangerFullAccess)
+        | (Never, DangerFullAccess)
+        | (OnRequest, DangerFullAccess) => SafetyCheck::AutoApprove {
             sandbox_type: SandboxType::None,
         },
+        (OnRequest, ReadOnly) | (OnRequest, WorkspaceWrite { .. }) => {
+            if with_escalated_permissions {
+                SafetyCheck::AskUser
+            } else {
+                match get_platform_sandbox() {
+                    Some(sandbox_type) => SafetyCheck::AutoApprove { sandbox_type },
+                    // Fall back to asking since the command is untrusted and
+                    // we do not have a sandbox available
+                    None => SafetyCheck::AskUser,
+                }
+            }
+        }
         (Never, ReadOnly)
         | (Never, WorkspaceWrite { .. })
         | (OnFailure, ReadOnly)
@@ -264,4 +280,47 @@ mod tests {
             &cwd,
         ))
     }
+
+    #[test]
+    fn test_request_escalated_privileges() {
+        // Should not be a trusted command
+        let command = vec!["git commit".to_string()];
+        let approval_policy = AskForApproval::OnRequest;
+        let sandbox_policy = SandboxPolicy::ReadOnly;
+        let approved: HashSet<Vec<String>> = HashSet::new();
+        let request_escalated_privileges = true;
+
+        let safety_check = assess_command_safety(
+            &command,
+            approval_policy,
+            &sandbox_policy,
+            &approved,
+            request_escalated_privileges,
+        );
+
+        assert_eq!(safety_check, SafetyCheck::AskUser);
+    }
+
+    #[test]
+    fn test_request_escalated_privileges_no_sandbox_fallback() {
+        let command = vec!["git".to_string(), "commit".to_string()];
+        let approval_policy = AskForApproval::OnRequest;
+        let sandbox_policy = SandboxPolicy::ReadOnly;
+        let approved: HashSet<Vec<String>> = HashSet::new();
+        let request_escalated_privileges = false;
+
+        let safety_check = assess_command_safety(
+            &command,
+            approval_policy,
+            &sandbox_policy,
+            &approved,
+            request_escalated_privileges,
+        );
+
+        let expected = match get_platform_sandbox() {
+            Some(sandbox_type) => SafetyCheck::AutoApprove { sandbox_type },
+            None => SafetyCheck::AskUser,
+        };
+        assert_eq!(safety_check, expected);
+    }
 }

codex-rs/core/src/shell.rs

@@ -215,6 +215,8 @@ mod tests {
                         "HOME".to_string(),
                         temp_home.path().to_str().unwrap().to_string(),
                     )]),
+                    with_escalated_permissions: None,
+                    justification: None,
                 },
                 SandboxType::None,
                 Arc::new(Notify::new()),

codex-rs/core/src/turn_diff_tracker.rs

@@ -546,6 +546,7 @@ index {ZERO_OID}..{right_oid}
             FileChange::Update {
                 unified_diff: "".to_owned(),
                 move_path: None,
+                hunks: None,
             },
         )]);
         acc.on_patch_begin(&update_changes);
@@ -616,6 +617,7 @@ index {left_oid}..{ZERO_OID}
             FileChange::Update {
                 unified_diff: "".to_owned(),
                 move_path: Some(dest.clone()),
+                hunks: None,
             },
         )]);
         acc.on_patch_begin(&mv_changes);
@@ -656,6 +658,7 @@ index {left_oid}..{right_oid}
             FileChange::Update {
                 unified_diff: "".to_owned(),
                 move_path: Some(dest.clone()),
+                hunks: None,
             },
         )]);
         acc.on_patch_begin(&mv_changes);
@@ -678,6 +681,7 @@ index {left_oid}..{right_oid}
             FileChange::Update {
                 unified_diff: "".into(),
                 move_path: Some(dest.clone()),
+                hunks: None,
             },
         )]);
         acc.on_patch_begin(&mv);
@@ -718,6 +722,7 @@ index {ZERO_OID}..{right_oid}
             FileChange::Update {
                 unified_diff: "".to_owned(),
                 move_path: None,
+                hunks: None,
             },
         )]);
         acc.on_patch_begin(&update_a);
@@ -793,6 +798,7 @@ index {left_oid_b}..{ZERO_OID}
             FileChange::Update {
                 unified_diff: "".to_owned(),
                 move_path: None,
+                hunks: None,
             },
         )]);
         acc.on_patch_begin(&update_changes);
@@ -859,6 +865,7 @@ index {ZERO_OID}..{right_oid}
             FileChange::Update {
                 unified_diff: "".to_owned(),
                 move_path: None,
+                hunks: None,
             },
         )]);
         acc.on_patch_begin(&update_changes);

codex-rs/core/tests/client.rs

@@ -1,3 +1,5 @@
+#![allow(clippy::expect_used)]
+#![allow(clippy::unwrap_used)]
 use std::path::PathBuf;
 
 use chrono::Utc;
@@ -32,6 +34,32 @@ fn sse_completed(id: &str) -> String {
     load_sse_fixture_with_id("tests/fixtures/completed_template.json", id)
 }
 
+fn assert_message_role(request_body: &serde_json::Value, role: &str) {
+    assert_eq!(request_body["role"].as_str().unwrap(), role);
+}
+
+fn assert_message_starts_with(request_body: &serde_json::Value, text: &str) {
+    let content = request_body["content"][0]["text"]
+        .as_str()
+        .expect("invalid message content");
+
+    assert!(
+        content.starts_with(text),
+        "expected message content '{content}' to start with '{text}'"
+    );
+}
+
+fn assert_message_ends_with(request_body: &serde_json::Value, text: &str) {
+    let content = request_body["content"][0]["text"]
+        .as_str()
+        .expect("invalid message content");
+
+    assert!(
+        content.ends_with(text),
+        "expected message content '{content}' to end with '{text}'"
+    );
+}
+
 #[tokio::test(flavor = "multi_thread", worker_threads = 2)]
 async fn includes_session_id_and_model_headers_in_request() {
     #![allow(clippy::unwrap_used)]
@@ -371,19 +399,12 @@ async fn includes_user_instructions_message_in_request() {
             .unwrap()
             .contains("be nice")
     );
-    assert_eq!(request_body["input"][0]["role"], "user");
-    assert!(
-        request_body["input"][0]["content"][0]["text"]
-            .as_str()
-            .unwrap()
-            .starts_with("<user_instructions>\n\nbe nice")
-    );
-    assert!(
-        request_body["input"][0]["content"][0]["text"]
-            .as_str()
-            .unwrap()
-            .ends_with("</user_instructions>")
-    );
+    assert_message_role(&request_body["input"][0], "user");
+    assert_message_starts_with(&request_body["input"][0], "<environment_context>\n\n");
+    assert_message_ends_with(&request_body["input"][0], "</environment_context>");
+    assert_message_role(&request_body["input"][1], "user");
+    assert_message_starts_with(&request_body["input"][1], "<user_instructions>\n\n");
+    assert_message_ends_with(&request_body["input"][1], "</user_instructions>");
 }
 
 #[tokio::test(flavor = "multi_thread", worker_threads = 2)]
@@ -437,7 +458,7 @@ async fn azure_overrides_assign_properties_used_for_responses_url() {
         request_max_retries: None,
         stream_max_retries: None,
         stream_idle_timeout_ms: None,
-        requires_auth: false,
+        requires_openai_auth: false,
     };
 
     // Init session
@@ -460,6 +481,86 @@ async fn azure_overrides_assign_properties_used_for_responses_url() {
     wait_for_event(&codex, |ev| matches!(ev, EventMsg::TaskComplete(_))).await;
 }
 
+#[tokio::test(flavor = "multi_thread", worker_threads = 2)]
+async fn env_var_overrides_loaded_auth() {
+    #![allow(clippy::unwrap_used)]
+
+    let existing_env_var_with_random_value = if cfg!(windows) { "USERNAME" } else { "USER" };
+
+    // Mock server
+    let server = MockServer::start().await;
+
+    // First request – must NOT include `previous_response_id`.
+    let first = ResponseTemplate::new(200)
+        .insert_header("content-type", "text/event-stream")
+        .set_body_raw(sse_completed("resp1"), "text/event-stream");
+
+    // Expect POST to /openai/responses with api-version query param
+    Mock::given(method("POST"))
+        .and(path("/openai/responses"))
+        .and(query_param("api-version", "2025-04-01-preview"))
+        .and(header_regex("Custom-Header", "Value"))
+        .and(header_regex(
+            "Authorization",
+            format!(
+                "Bearer {}",
+                std::env::var(existing_env_var_with_random_value).unwrap()
+            )
+            .as_str(),
+        ))
+        .respond_with(first)
+        .expect(1)
+        .mount(&server)
+        .await;
+
+    let provider = ModelProviderInfo {
+        name: "custom".to_string(),
+        base_url: Some(format!("{}/openai", server.uri())),
+        // Reuse the existing environment variable to avoid using unsafe code
+        env_key: Some(existing_env_var_with_random_value.to_string()),
+        query_params: Some(std::collections::HashMap::from([(
+            "api-version".to_string(),
+            "2025-04-01-preview".to_string(),
+        )])),
+        env_key_instructions: None,
+        wire_api: WireApi::Responses,
+        http_headers: Some(std::collections::HashMap::from([(
+            "Custom-Header".to_string(),
+            "Value".to_string(),
+        )])),
+        env_http_headers: None,
+        request_max_retries: None,
+        stream_max_retries: None,
+        stream_idle_timeout_ms: None,
+        requires_openai_auth: false,
+    };
+
+    // Init session
+    let codex_home = TempDir::new().unwrap();
+    let mut config = load_default_config_for_test(&codex_home);
+    config.model_provider = provider;
+
+    let ctrl_c = std::sync::Arc::new(tokio::sync::Notify::new());
+    let CodexSpawnOk { codex, .. } = Codex::spawn(
+        config,
+        Some(auth_from_token("Default Access Token".to_string())),
+        ctrl_c.clone(),
+    )
+    .await
+    .unwrap();
+
+    codex
+        .submit(Op::UserInput {
+            items: vec![InputItem::Text {
+                text: "hello".into(),
+            }],
+        })
+        .await
+        .unwrap();
+
+    wait_for_event(&codex, |ev| matches!(ev, EventMsg::TaskComplete(_))).await;
+}
+
 fn auth_from_token(id_token: String) -> CodexAuth {
     CodexAuth::new(
         None,

codex-rs/core/tests/exec.rs

@@ -28,6 +28,8 @@ async fn run_test_cmd(tmp: TempDir, cmd: Vec<&str>, should_be_ok: bool) {
         cwd: tmp.path().to_path_buf(),
         timeout_ms: Some(1000),
         env: HashMap::new(),
+        with_escalated_permissions: None,
+        justification: None,
     };
 
     let ctrl_c = Arc::new(Notify::new());

codex-rs/core/tests/exec_stream_events.rs

@@ -53,6 +53,8 @@ async fn test_exec_stdout_stream_events_echo() {
         cwd: std::env::current_dir().unwrap_or_else(|_| PathBuf::from(".")),
         timeout_ms: Some(5_000),
         env: HashMap::new(),
+        with_escalated_permissions: None,
+        justification: None,
     };
 
     let ctrl_c = Arc::new(Notify::new());
@@ -103,6 +105,8 @@ async fn test_exec_stderr_stream_events_echo() {
         cwd: std::env::current_dir().unwrap_or_else(|_| PathBuf::from(".")),
         timeout_ms: Some(5_000),
         env: HashMap::new(),
+        with_escalated_permissions: None,
+        justification: None,
     };
 
     let ctrl_c = Arc::new(Notify::new());

codex-rs/core/tests/stream_no_completed.rs

@@ -90,7 +90,7 @@ async fn retries_on_early_close() {
         request_max_retries: Some(0),
         stream_max_retries: Some(1),
         stream_idle_timeout_ms: Some(2000),
-        requires_auth: false,
+        requires_openai_auth: false,
     };
 
     let ctrl_c = std::sync::Arc::new(tokio::sync::Notify::new());

codex-rs/exec/Cargo.toml

@@ -25,6 +25,7 @@ codex-common = { path = "../common", features = [
     "sandbox_summary",
 ] }
 codex-core = { path = "../core" }
+codex-ollama = { path = "../ollama" }
 owo-colors = "4.2.0"
 serde_json = "1"
 shlex = "1.3.0"

codex-rs/exec/src/cli.rs

@@ -14,6 +14,9 @@ pub struct Cli {
     #[arg(long, short = 'm')]
     pub model: Option<String>,
 
+    #[arg(long = "oss", default_value_t = false)]
+    pub oss: bool,
+
     /// Select the sandbox policy to use when executing model-generated shell
     /// commands.
     #[arg(long = "sandbox", short = 's')]

codex-rs/exec/src/event_processor.rs

@@ -1,9 +1,6 @@
 use std::path::Path;
 
-use codex_common::summarize_sandbox_policy;
-use codex_core::WireApi;
 use codex_core::config::Config;
-use codex_core::model_supports_reasoning_summaries;
 use codex_core::protocol::Event;
 
 pub(crate) enum CodexStatus {
@@ -20,30 +17,6 @@ pub(crate) trait EventProcessor {
     fn process_event(&mut self, event: Event) -> CodexStatus;
 }
 
-pub(crate) fn create_config_summary_entries(config: &Config) -> Vec<(&'static str, String)> {
-    let mut entries = vec![
-        ("workdir", config.cwd.display().to_string()),
-        ("model", config.model.clone()),
-        ("provider", config.model_provider_id.clone()),
-        ("approval", config.approval_policy.to_string()),
-        ("sandbox", summarize_sandbox_policy(&config.sandbox_policy)),
-    ];
-    if config.model_provider.wire_api == WireApi::Responses
-        && model_supports_reasoning_summaries(config)
-    {
-        entries.push((
-            "reasoning effort",
-            config.model_reasoning_effort.to_string(),
-        ));
-        entries.push((
-            "reasoning summaries",
-            config.model_reasoning_summary.to_string(),
-        ));
-    }
-
-    entries
-}
-
 pub(crate) fn handle_last_message(last_agent_message: Option<&str>, output_file: &Path) {
     let message = last_agent_message.unwrap_or_default();
     write_last_message_file(message, Some(output_file));

codex-rs/exec/src/event_processor_with_human_output.rs

@@ -5,6 +5,8 @@ use codex_core::plan_tool::UpdatePlanArgs;
 use codex_core::protocol::AgentMessageDeltaEvent;
 use codex_core::protocol::AgentMessageEvent;
 use codex_core::protocol::AgentReasoningDeltaEvent;
+use codex_core::protocol::AgentReasoningRawContentDeltaEvent;
+use codex_core::protocol::AgentReasoningRawContentEvent;
 use codex_core::protocol::BackgroundEventEvent;
 use codex_core::protocol::ErrorEvent;
 use codex_core::protocol::Event;
@@ -31,8 +33,8 @@ use std::time::Instant;
 
 use crate::event_processor::CodexStatus;
 use crate::event_processor::EventProcessor;
-use crate::event_processor::create_config_summary_entries;
 use crate::event_processor::handle_last_message;
+use codex_common::create_config_summary_entries;
 
 /// This should be configurable. When used in CI, users may not want to impose
 /// a limit so they can see the full transcript.
@@ -55,8 +57,10 @@ pub(crate) struct EventProcessorWithHumanOutput {
 
     /// Whether to include `AgentReasoning` events in the output.
     show_agent_reasoning: bool,
+    show_raw_agent_reasoning: bool,
     answer_started: bool,
     reasoning_started: bool,
+    raw_reasoning_started: bool,
     last_message_path: Option<PathBuf>,
 }
 
@@ -81,8 +85,10 @@ impl EventProcessorWithHumanOutput {
                 green: Style::new().green(),
                 cyan: Style::new().cyan(),
                 show_agent_reasoning: !config.hide_agent_reasoning,
+                show_raw_agent_reasoning: config.show_raw_agent_reasoning,
                 answer_started: false,
                 reasoning_started: false,
+                raw_reasoning_started: false,
                 last_message_path,
             }
         } else {
@@ -97,8 +103,10 @@ impl EventProcessorWithHumanOutput {
                 green: Style::new(),
                 cyan: Style::new(),
                 show_agent_reasoning: !config.hide_agent_reasoning,
+                show_raw_agent_reasoning: config.show_raw_agent_reasoning,
                 answer_started: false,
                 reasoning_started: false,
+                raw_reasoning_started: false,
                 last_message_path,
             }
         }
@@ -203,6 +211,32 @@ impl EventProcessor for EventProcessorWithHumanOutput {
                 #[allow(clippy::expect_used)]
                 std::io::stdout().flush().expect("could not flush stdout");
             }
+            EventMsg::AgentReasoningRawContent(AgentReasoningRawContentEvent { text }) => {
+                if !self.show_raw_agent_reasoning {
+                    return CodexStatus::Running;
+                }
+                if !self.raw_reasoning_started {
+                    print!("{text}");
+                    #[allow(clippy::expect_used)]
+                    std::io::stdout().flush().expect("could not flush stdout");
+                } else {
+                    println!();
+                    self.raw_reasoning_started = false;
+                }
+            }
+            EventMsg::AgentReasoningRawContentDelta(AgentReasoningRawContentDeltaEvent {
+                delta,
+            }) => {
+                if !self.show_raw_agent_reasoning {
+                    return CodexStatus::Running;
+                }
+                if !self.raw_reasoning_started {
+                    self.raw_reasoning_started = true;
+                }
+                print!("{delta}");
+                #[allow(clippy::expect_used)]
+                std::io::stdout().flush().expect("could not flush stdout");
+            }
             EventMsg::AgentMessage(AgentMessageEvent { message }) => {
                 // if answer_started is false, this means we haven't received any
                 // delta. Thus, we need to print the message as a new answer.
@@ -363,6 +397,7 @@ impl EventProcessor for EventProcessorWithHumanOutput {
                         FileChange::Update {
                             unified_diff,
                             move_path,
+                            ..
                         } => {
                             let header = if let Some(dest) = move_path {
                                 format!(

codex-rs/exec/src/event_processor_with_json_output.rs

@@ -9,8 +9,8 @@ use serde_json::json;
 
 use crate::event_processor::CodexStatus;
 use crate::event_processor::EventProcessor;
-use crate::event_processor::create_config_summary_entries;
 use crate::event_processor::handle_last_message;
+use codex_common::create_config_summary_entries;
 
 pub(crate) struct EventProcessorWithJsonOutput {
     last_message_path: Option<PathBuf>,

codex-rs/exec/src/lib.rs

@@ -9,6 +9,7 @@ use std::path::PathBuf;
 use std::sync::Arc;
 
 pub use cli::Cli;
+use codex_core::BUILT_IN_OSS_MODEL_PROVIDER_ID;
 use codex_core::codex_wrapper::CodexConversation;
 use codex_core::codex_wrapper::{self};
 use codex_core::config::Config;
@@ -21,6 +22,7 @@ use codex_core::protocol::InputItem;
 use codex_core::protocol::Op;
 use codex_core::protocol::TaskCompleteEvent;
 use codex_core::util::is_inside_git_repo;
+use codex_ollama::DEFAULT_OSS_MODEL;
 use event_processor_with_human_output::EventProcessorWithHumanOutput;
 use event_processor_with_json_output::EventProcessorWithJsonOutput;
 use tracing::debug;
@@ -34,7 +36,8 @@ use crate::event_processor::EventProcessor;
 pub async fn run_main(cli: Cli, codex_linux_sandbox_exe: Option<PathBuf>) -> anyhow::Result<()> {
     let Cli {
         images,
-        model,
+        model: model_cli_arg,
+        oss,
         config_profile,
         full_auto,
         dangerously_bypass_approvals_and_sandbox,
@@ -114,6 +117,23 @@ pub async fn run_main(cli: Cli, codex_linux_sandbox_exe: Option<PathBuf>) -> any
         sandbox_mode_cli_arg.map(Into::<SandboxMode>::into)
     };
 
+    // When using `--oss`, let the bootstrapper pick the model (defaulting to
+    // gpt-oss:20b) and ensure it is present locally. Also, force the built‑in
+    // `oss` model provider.
+    let model = if let Some(model) = model_cli_arg {
+        Some(model)
+    } else if oss {
+        Some(DEFAULT_OSS_MODEL.to_owned())
+    } else {
+        None // No model specified, will use the default.
+    };
+
+    let model_provider = if oss {
+        Some(BUILT_IN_OSS_MODEL_PROVIDER_ID.to_string())
+    } else {
+        None // No specific model provider override.
+    };
+
     // Load configuration and determine approval policy
     let overrides = ConfigOverrides {
         model,
@@ -123,10 +143,12 @@ pub async fn run_main(cli: Cli, codex_linux_sandbox_exe: Option<PathBuf>) -> any
         approval_policy: Some(AskForApproval::Never),
         sandbox_mode,
         cwd: cwd.map(|p| p.canonicalize().unwrap_or(p)),
-        model_provider: None,
+        model_provider,
         codex_linux_sandbox_exe,
         base_instructions: None,
         include_plan_tool: None,
+        disable_response_storage: oss.then_some(true),
+        show_raw_agent_reasoning: oss.then_some(true),
     };
     // Parse `-c` overrides.
     let cli_kv_overrides = match config_overrides.parse_overrides() {
@@ -148,6 +170,12 @@ pub async fn run_main(cli: Cli, codex_linux_sandbox_exe: Option<PathBuf>) -> any
         ))
     };
 
+    if oss {
+        codex_ollama::ensure_oss_ready(&config)
+            .await
+            .map_err(|e| anyhow::anyhow!("OSS setup failed: {e}"))?;
+    }
+
     // Print the effective configuration and prompt so users can see what Codex
     // is using.
     event_processor.print_config_summary(&config, &prompt);
@@ -188,10 +216,16 @@ pub async fn run_main(cli: Cli, codex_linux_sandbox_exe: Option<PathBuf>) -> any
                     res = codex.next_event() => match res {
                         Ok(event) => {
                             debug!("Received event: {event:?}");
+
+                            let is_shutdown_complete = matches!(event.msg, EventMsg::ShutdownComplete);
                             if let Err(e) = tx.send(event) {
                                 error!("Error sending event: {e:?}");
                                 break;
                             }
+                            if is_shutdown_complete {
+                                info!("Received shutdown event, exiting event loop.");
+                                break;
+                            }
                         },
                         Err(e) => {
                             error!("Error receiving event: {e:?}");

codex-rs/linux-sandbox/tests/landlock.rs

@@ -44,6 +44,8 @@ async fn run_cmd(cmd: &[&str], writable_roots: &[PathBuf], timeout_ms: u64) {
         cwd: std::env::current_dir().expect("cwd should exist"),
         timeout_ms: Some(timeout_ms),
         env: create_env_from_core_vars(),
+        with_escalated_permissions: None,
+        justification: None,
     };
 
     let sandbox_policy = SandboxPolicy::WorkspaceWrite {
@@ -139,6 +141,8 @@ async fn assert_network_blocked(cmd: &[&str]) {
         // do not stall the suite.
         timeout_ms: Some(NETWORK_TIMEOUT_MS),
         env: create_env_from_core_vars(),
+        with_escalated_permissions: None,
+        justification: None,
     };
 
     let sandbox_policy = SandboxPolicy::new_read_only_policy();

codex-rs/login/src/lib.rs

@@ -4,6 +4,7 @@ use chrono::Utc;
 use serde::Deserialize;
 use serde::Serialize;
 use std::env;
+use std::fs::File;
 use std::fs::OpenOptions;
 use std::io::Read;
 use std::io::Write;
@@ -11,6 +12,7 @@ use std::io::Write;
 use std::os::unix::fs::OpenOptionsExt;
 use std::path::Path;
 use std::path::PathBuf;
+use std::process::Child;
 use std::process::Stdio;
 use std::sync::Arc;
 use std::sync::Mutex;
@@ -183,6 +185,59 @@ fn get_auth_file(codex_home: &Path) -> PathBuf {
     codex_home.join("auth.json")
 }
 
+/// Represents a running login subprocess. The child can be killed by holding
+/// the mutex and calling `kill()`.
+#[derive(Debug, Clone)]
+pub struct SpawnedLogin {
+    pub child: Arc<Mutex<Child>>,
+    pub stdout: Arc<Mutex<Vec<u8>>>,
+    pub stderr: Arc<Mutex<Vec<u8>>>,
+}
+
+/// Spawn the ChatGPT login Python server as a child process and return a handle to its process.
+pub fn spawn_login_with_chatgpt(codex_home: &Path) -> std::io::Result<SpawnedLogin> {
+    let mut cmd = std::process::Command::new("python3");
+    cmd.arg("-c")
+        .arg(SOURCE_FOR_PYTHON_SERVER)
+        .env("CODEX_HOME", codex_home)
+        .env("CODEX_CLIENT_ID", CLIENT_ID)
+        .stdin(Stdio::null())
+        .stdout(Stdio::piped())
+        .stderr(Stdio::piped());
+
+    let mut child = cmd.spawn()?;
+
+    let stdout_buf = Arc::new(Mutex::new(Vec::new()));
+    let stderr_buf = Arc::new(Mutex::new(Vec::new()));
+
+    if let Some(mut out) = child.stdout.take() {
+        let buf = stdout_buf.clone();
+        std::thread::spawn(move || {
+            let mut tmp = Vec::new();
+            let _ = std::io::copy(&mut out, &mut tmp);
+            if let Ok(mut b) = buf.lock() {
+                b.extend_from_slice(&tmp);
+            }
+        });
+    }
+    if let Some(mut err) = child.stderr.take() {
+        let buf = stderr_buf.clone();
+        std::thread::spawn(move || {
+            let mut tmp = Vec::new();
+            let _ = std::io::copy(&mut err, &mut tmp);
+            if let Ok(mut b) = buf.lock() {
+                b.extend_from_slice(&tmp);
+            }
+        });
+    }
+
+    Ok(SpawnedLogin {
+        child: Arc::new(Mutex::new(child)),
+        stdout: stdout_buf,
+        stderr: stderr_buf,
+    })
+}
+
 /// Run `python3 -c {{SOURCE_FOR_PYTHON_SERVER}}` with the CODEX_HOME
 /// environment variable set to the provided `codex_home` path. If the
 /// subprocess exits 0, read the OPENAI_API_KEY property out of
@@ -234,7 +289,7 @@ pub fn login_with_api_key(codex_home: &Path, api_key: &str) -> std::io::Result<(
 /// Attempt to read and refresh the `auth.json` file in the given `CODEX_HOME` directory.
 /// Returns the full AuthDotJson structure after refreshing if necessary.
 pub fn try_read_auth_json(auth_file: &Path) -> std::io::Result<AuthDotJson> {
-    let mut file = std::fs::File::open(auth_file)?;
+    let mut file = File::open(auth_file)?;
     let mut contents = String::new();
     file.read_to_string(&mut contents)?;
     let auth_dot_json: AuthDotJson = serde_json::from_str(&contents)?;

codex-rs/login/src/login_with_chatgpt.py

@@ -110,7 +110,7 @@ def main() -> None:
                 eprint(f"Failed to open browser: {e}")
 
         eprint(
-            f"If your browser did not open, navigate to this URL to authenticate:\n\n{auth_url}"
+            f". If your browser did not open, navigate to this URL to authenticate: \n\n{auth_url}"
         )
 
         # Run the server in the main thread until `shutdown()` is called by the

codex-rs/mcp-server/src/codex_tool_config.rs

@@ -158,6 +158,8 @@ impl CodexToolCallParam {
             codex_linux_sandbox_exe,
             base_instructions,
             include_plan_tool,
+            disable_response_storage: None,
+            show_raw_agent_reasoning: None,
         };
 
         let cli_overrides = cli_overrides

codex-rs/mcp-server/src/codex_tool_runner.rs

@@ -252,7 +252,9 @@ async fn run_codex_tool_session_inner(
                     EventMsg::AgentMessage(AgentMessageEvent { .. }) => {
                         // TODO: think how we want to support this in the MCP
                     }
-                    EventMsg::TaskStarted
+                    EventMsg::AgentReasoningRawContent(_)
+                    | EventMsg::AgentReasoningRawContentDelta(_)
+                    | EventMsg::TaskStarted
                     | EventMsg::TokenCount(_)
                     | EventMsg::AgentReasoning(_)
                     | EventMsg::McpToolCallBegin(_)

codex-rs/mcp-server/src/conversation_loop.rs

@@ -90,7 +90,9 @@ pub async fn run_conversation_loop(
                     EventMsg::AgentMessage(AgentMessageEvent { .. }) => {
                         // TODO: think how we want to support this in the MCP
                     }
-                    EventMsg::TaskStarted
+                    EventMsg::AgentReasoningRawContent(_)
+                    | EventMsg::AgentReasoningRawContentDelta(_)
+                    | EventMsg::TaskStarted
                     | EventMsg::TokenCount(_)
                     | EventMsg::AgentReasoning(_)
                     | EventMsg::McpToolCallBegin(_)

codex-rs/mcp-server/src/tool_handlers/create_conversation.rs

@@ -59,6 +59,8 @@ pub(crate) async fn handle_create_conversation(
         codex_linux_sandbox_exe: None,
         base_instructions,
         include_plan_tool: None,
+        disable_response_storage: None,
+        show_raw_agent_reasoning: None,
     };
 
     let cfg: CodexConfig = match CodexConfig::load_with_cli_overrides(cli_overrides, overrides) {

codex-rs/mcp-server/tests/codex_tool.rs

@@ -236,6 +236,7 @@ async fn patch_approval_triggers_elicitation() -> anyhow::Result<()> {
         FileChange::Update {
             unified_diff: "@@ -1 +1 @@\n-original content\n+modified content\n".to_string(),
             move_path: None,
+            hunks: None,
         },
     );
 

codex-rs/mcp-server/tests/send_message.rs

@@ -18,7 +18,7 @@ use tokio::time::timeout;
 
 const DEFAULT_READ_TIMEOUT: std::time::Duration = std::time::Duration::from_secs(10);
 
-#[tokio::test(flavor = "multi_thread", worker_threads = 2)]
+#[tokio::test]
 async fn test_send_message_success() {
     // Spin up a mock completions server that immediately ends the Codex turn.
     // Two Codex turns hit the mock model (session start + send-user-message). Provide two SSE responses.
@@ -99,13 +99,13 @@ async fn test_send_message_success() {
         response
     );
     // wait for the server to hear the user message
-    sleep(Duration::from_secs(1));
+    sleep(Duration::from_secs(5));
 
     // Ensure the server and tempdir live until end of test
     drop(server);
 }
 
-#[tokio::test(flavor = "multi_thread", worker_threads = 2)]
+#[tokio::test]
 async fn test_send_message_session_not_found() {
     // Start MCP without creating a Codex session
     let codex_home = TempDir::new().expect("tempdir");

codex-rs/ollama/Cargo.toml

@@ -0,0 +1,32 @@
+[package]
+edition = "2024"
+name = "codex-ollama"
+version = { workspace = true }
+
+[lib]
+name = "codex_ollama"
+path = "src/lib.rs"
+
+[lints]
+workspace = true
+
+[dependencies]
+async-stream = "0.3"
+bytes = "1.10.1"
+codex-core = { path = "../core" }
+futures = "0.3"
+reqwest = { version = "0.12", features = ["json", "stream"] }
+serde_json = "1"
+tokio = { version = "1", features = [
+    "io-std",
+    "macros",
+    "process",
+    "rt-multi-thread",
+    "signal",
+] }
+toml = "0.9.2"
+tracing = { version = "0.1.41", features = ["log"] }
+wiremock = "0.6"
+
+[dev-dependencies]
+tempfile = "3"

codex-rs/ollama/src/client.rs

@@ -0,0 +1,351 @@
+use bytes::BytesMut;
+use futures::StreamExt;
+use futures::stream::BoxStream;
+use serde_json::Value as JsonValue;
+use std::collections::VecDeque;
+use std::io;
+
+use crate::parser::pull_events_from_value;
+use crate::pull::PullEvent;
+use crate::pull::PullProgressReporter;
+use crate::url::base_url_to_host_root;
+use crate::url::is_openai_compatible_base_url;
+use codex_core::BUILT_IN_OSS_MODEL_PROVIDER_ID;
+use codex_core::ModelProviderInfo;
+use codex_core::WireApi;
+use codex_core::config::Config;
+
+const OLLAMA_CONNECTION_ERROR: &str = "No running Ollama server detected. Start it with: `ollama serve` (after installing). Install instructions: https://github.com/ollama/ollama?tab=readme-ov-file#ollama";
+
+/// Client for interacting with a local Ollama instance.
+pub struct OllamaClient {
+    client: reqwest::Client,
+    host_root: String,
+    uses_openai_compat: bool,
+}
+
+impl OllamaClient {
+    /// Construct a client for the built‑in open‑source ("oss") model provider
+    /// and verify that a local Ollama server is reachable. If no server is
+    /// detected, returns an error with helpful installation/run instructions.
+    pub async fn try_from_oss_provider(config: &Config) -> io::Result<Self> {
+        // Note that we must look up the provider from the Config to ensure that
+        // any overrides the user has in their config.toml are taken into
+        // account.
+        let provider = config
+            .model_providers
+            .get(BUILT_IN_OSS_MODEL_PROVIDER_ID)
+            .ok_or_else(|| {
+                io::Error::new(
+                    io::ErrorKind::NotFound,
+                    format!("Built-in provider {BUILT_IN_OSS_MODEL_PROVIDER_ID} not found",),
+                )
+            })?;
+
+        Self::try_from_provider(provider).await
+    }
+
+    #[cfg(test)]
+    async fn try_from_provider_with_base_url(base_url: &str) -> io::Result<Self> {
+        let provider = codex_core::create_oss_provider_with_base_url(base_url);
+        Self::try_from_provider(&provider).await
+    }
+
+    /// Build a client from a provider definition and verify the server is reachable.
+    async fn try_from_provider(provider: &ModelProviderInfo) -> io::Result<Self> {
+        #![allow(clippy::expect_used)]
+        let base_url = provider
+            .base_url
+            .as_ref()
+            .expect("oss provider must have a base_url");
+        let uses_openai_compat = is_openai_compatible_base_url(base_url)
+            || matches!(provider.wire_api, WireApi::Chat)
+                && is_openai_compatible_base_url(base_url);
+        let host_root = base_url_to_host_root(base_url);
+        let client = reqwest::Client::builder()
+            .connect_timeout(std::time::Duration::from_secs(5))
+            .build()
+            .unwrap_or_else(|_| reqwest::Client::new());
+        let client = Self {
+            client,
+            host_root,
+            uses_openai_compat,
+        };
+        client.probe_server().await?;
+        Ok(client)
+    }
+
+    /// Probe whether the server is reachable by hitting the appropriate health endpoint.
+    async fn probe_server(&self) -> io::Result<()> {
+        let url = if self.uses_openai_compat {
+            format!("{}/v1/models", self.host_root.trim_end_matches('/'))
+        } else {
+            format!("{}/api/tags", self.host_root.trim_end_matches('/'))
+        };
+        let resp = self.client.get(url).send().await.map_err(|err| {
+            tracing::warn!("Failed to connect to Ollama server: {err:?}");
+            io::Error::other(OLLAMA_CONNECTION_ERROR)
+        })?;
+        if resp.status().is_success() {
+            Ok(())
+        } else {
+            tracing::warn!(
+                "Failed to probe server at {}: HTTP {}",
+                self.host_root,
+                resp.status()
+            );
+            Err(io::Error::other(OLLAMA_CONNECTION_ERROR))
+        }
+    }
+
+    /// Return the list of model names known to the local Ollama instance.
+    pub async fn fetch_models(&self) -> io::Result<Vec<String>> {
+        let tags_url = format!("{}/api/tags", self.host_root.trim_end_matches('/'));
+        let resp = self
+            .client
+            .get(tags_url)
+            .send()
+            .await
+            .map_err(io::Error::other)?;
+        if !resp.status().is_success() {
+            return Ok(Vec::new());
+        }
+        let val = resp.json::<JsonValue>().await.map_err(io::Error::other)?;
+        let names = val
+            .get("models")
+            .and_then(|m| m.as_array())
+            .map(|arr| {
+                arr.iter()
+                    .filter_map(|v| v.get("name").and_then(|n| n.as_str()))
+                    .map(|s| s.to_string())
+                    .collect::<Vec<_>>()
+            })
+            .unwrap_or_default();
+        Ok(names)
+    }
+
+    /// Start a model pull and emit streaming events. The returned stream ends when
+    /// a Success event is observed or the server closes the connection.
+    pub async fn pull_model_stream(
+        &self,
+        model: &str,
+    ) -> io::Result<BoxStream<'static, PullEvent>> {
+        let url = format!("{}/api/pull", self.host_root.trim_end_matches('/'));
+        let resp = self
+            .client
+            .post(url)
+            .json(&serde_json::json!({"model": model, "stream": true}))
+            .send()
+            .await
+            .map_err(io::Error::other)?;
+        if !resp.status().is_success() {
+            return Err(io::Error::other(format!(
+                "failed to start pull: HTTP {}",
+                resp.status()
+            )));
+        }
+
+        let mut stream = resp.bytes_stream();
+        let mut buf = BytesMut::new();
+        let _pending: VecDeque<PullEvent> = VecDeque::new();
+
+        // Using an async stream adaptor backed by unfold-like manual loop.
+        let s = async_stream::stream! {
+            while let Some(chunk) = stream.next().await {
+                match chunk {
+                    Ok(bytes) => {
+                        buf.extend_from_slice(&bytes);
+                        while let Some(pos) = buf.iter().position(|b| *b == b'\n') {
+                            let line = buf.split_to(pos + 1);
+                            if let Ok(text) = std::str::from_utf8(&line) {
+                                let text = text.trim();
+                                if text.is_empty() { continue; }
+                                if let Ok(value) = serde_json::from_str::<JsonValue>(text) {
+                                    for ev in pull_events_from_value(&value) { yield ev; }
+                                    if let Some(err_msg) = value.get("error").and_then(|e| e.as_str()) {
+                                        yield PullEvent::Error(err_msg.to_string());
+                                        return;
+                                    }
+                                    if let Some(status) = value.get("status").and_then(|s| s.as_str()) {
+                                        if status == "success" { yield PullEvent::Success; return; }
+                                    }
+                                }
+                            }
+                        }
+                    }
+                    Err(_) => {
+                        // Connection error: end the stream.
+                        return;
+                    }
+                }
+            }
+        };
+
+        Ok(Box::pin(s))
+    }
+
+    /// High-level helper to pull a model and drive a progress reporter.
+    pub async fn pull_with_reporter(
+        &self,
+        model: &str,
+        reporter: &mut dyn PullProgressReporter,
+    ) -> io::Result<()> {
+        reporter.on_event(&PullEvent::Status(format!("Pulling model {model}...")))?;
+        let mut stream = self.pull_model_stream(model).await?;
+        while let Some(event) = stream.next().await {
+            reporter.on_event(&event)?;
+            match event {
+                PullEvent::Success => {
+                    return Ok(());
+                }
+                PullEvent::Error(err) => {
+                    // Empirically, ollama returns a 200 OK response even when
+                    // the output stream includes an error message. Verify with:
+                    //
+                    // `curl -i http://localhost:11434/api/pull -d '{ "model": "foobarbaz" }'`
+                    //
+                    // As such, we have to check the event stream, not the
+                    // HTTP response status, to determine whether to return Err.
+                    return Err(io::Error::other(format!("Pull failed: {err}")));
+                }
+                PullEvent::ChunkProgress { .. } | PullEvent::Status(_) => {
+                    continue;
+                }
+            }
+        }
+        Err(io::Error::other(
+            "Pull stream ended unexpectedly without success.",
+        ))
+    }
+
+    /// Low-level constructor given a raw host root, e.g. "http://localhost:11434".
+    #[cfg(test)]
+    fn from_host_root(host_root: impl Into<String>) -> Self {
+        let client = reqwest::Client::builder()
+            .connect_timeout(std::time::Duration::from_secs(5))
+            .build()
+            .unwrap_or_else(|_| reqwest::Client::new());
+        Self {
+            client,
+            host_root: host_root.into(),
+            uses_openai_compat: false,
+        }
+    }
+}
+
+#[cfg(test)]
+mod tests {
+    #![allow(clippy::expect_used, clippy::unwrap_used)]
+    use super::*;
+
+    // Happy-path tests using a mock HTTP server; skip if sandbox network is disabled.
+    #[tokio::test]
+    async fn test_fetch_models_happy_path() {
+        if std::env::var(codex_core::spawn::CODEX_SANDBOX_NETWORK_DISABLED_ENV_VAR).is_ok() {
+            tracing::info!(
+                "{} is set; skipping test_fetch_models_happy_path",
+                codex_core::spawn::CODEX_SANDBOX_NETWORK_DISABLED_ENV_VAR
+            );
+            return;
+        }
+
+        let server = wiremock::MockServer::start().await;
+        wiremock::Mock::given(wiremock::matchers::method("GET"))
+            .and(wiremock::matchers::path("/api/tags"))
+            .respond_with(
+                wiremock::ResponseTemplate::new(200).set_body_raw(
+                    serde_json::json!({
+                        "models": [ {"name": "llama3.2:3b"}, {"name":"mistral"} ]
+                    })
+                    .to_string(),
+                    "application/json",
+                ),
+            )
+            .mount(&server)
+            .await;
+
+        let client = OllamaClient::from_host_root(server.uri());
+        let models = client.fetch_models().await.expect("fetch models");
+        assert!(models.contains(&"llama3.2:3b".to_string()));
+        assert!(models.contains(&"mistral".to_string()));
+    }
+
+    #[tokio::test]
+    async fn test_probe_server_happy_path_openai_compat_and_native() {
+        if std::env::var(codex_core::spawn::CODEX_SANDBOX_NETWORK_DISABLED_ENV_VAR).is_ok() {
+            tracing::info!(
+                "{} set; skipping test_probe_server_happy_path_openai_compat_and_native",
+                codex_core::spawn::CODEX_SANDBOX_NETWORK_DISABLED_ENV_VAR
+            );
+            return;
+        }
+
+        let server = wiremock::MockServer::start().await;
+
+        // Native endpoint
+        wiremock::Mock::given(wiremock::matchers::method("GET"))
+            .and(wiremock::matchers::path("/api/tags"))
+            .respond_with(wiremock::ResponseTemplate::new(200))
+            .mount(&server)
+            .await;
+        let native = OllamaClient::from_host_root(server.uri());
+        native.probe_server().await.expect("probe native");
+
+        // OpenAI compatibility endpoint
+        wiremock::Mock::given(wiremock::matchers::method("GET"))
+            .and(wiremock::matchers::path("/v1/models"))
+            .respond_with(wiremock::ResponseTemplate::new(200))
+            .mount(&server)
+            .await;
+        let ollama_client =
+            OllamaClient::try_from_provider_with_base_url(&format!("{}/v1", server.uri()))
+                .await
+                .expect("probe OpenAI compat");
+        ollama_client
+            .probe_server()
+            .await
+            .expect("probe OpenAI compat");
+    }
+
+    #[tokio::test]
+    async fn test_try_from_oss_provider_ok_when_server_running() {
+        if std::env::var(codex_core::spawn::CODEX_SANDBOX_NETWORK_DISABLED_ENV_VAR).is_ok() {
+            tracing::info!(
+                "{} set; skipping test_try_from_oss_provider_ok_when_server_running",
+                codex_core::spawn::CODEX_SANDBOX_NETWORK_DISABLED_ENV_VAR
+            );
+            return;
+        }
+
+        let server = wiremock::MockServer::start().await;
+
+        // OpenAI‑compat models endpoint responds OK.
+        wiremock::Mock::given(wiremock::matchers::method("GET"))
+            .and(wiremock::matchers::path("/v1/models"))
+            .respond_with(wiremock::ResponseTemplate::new(200))
+            .mount(&server)
+            .await;
+
+        OllamaClient::try_from_provider_with_base_url(&format!("{}/v1", server.uri()))
+            .await
+            .expect("client should be created when probe succeeds");
+    }
+
+    #[tokio::test]
+    async fn test_try_from_oss_provider_err_when_server_missing() {
+        if std::env::var(codex_core::spawn::CODEX_SANDBOX_NETWORK_DISABLED_ENV_VAR).is_ok() {
+            tracing::info!(
+                "{} set; skipping test_try_from_oss_provider_err_when_server_missing",
+                codex_core::spawn::CODEX_SANDBOX_NETWORK_DISABLED_ENV_VAR
+            );
+            return;
+        }
+
+        let server = wiremock::MockServer::start().await;
+        let err = OllamaClient::try_from_provider_with_base_url(&format!("{}/v1", server.uri()))
+            .await
+            .err()
+            .expect("expected error");
+        assert_eq!(OLLAMA_CONNECTION_ERROR, err.to_string());
+    }
+}

codex-rs/ollama/src/lib.rs

@@ -0,0 +1,44 @@
+mod client;
+mod parser;
+mod pull;
+mod url;
+
+pub use client::OllamaClient;
+use codex_core::config::Config;
+pub use pull::CliProgressReporter;
+pub use pull::PullEvent;
+pub use pull::PullProgressReporter;
+pub use pull::TuiProgressReporter;
+
+/// Default OSS model to use when `--oss` is passed without an explicit `-m`.
+pub const DEFAULT_OSS_MODEL: &str = "gpt-oss:20b";
+
+/// Prepare the local OSS environment when `--oss` is selected.
+///
+/// - Ensures a local Ollama server is reachable.
+/// - Checks if the model exists locally and pulls it if missing.
+pub async fn ensure_oss_ready(config: &Config) -> std::io::Result<()> {
+    // Only download when the requested model is the default OSS model (or when -m is not provided).
+    let model = config.model.as_ref();
+
+    // Verify local Ollama is reachable.
+    let ollama_client = crate::OllamaClient::try_from_oss_provider(config).await?;
+
+    // If the model is not present locally, pull it.
+    match ollama_client.fetch_models().await {
+        Ok(models) => {
+            if !models.iter().any(|m| m == model) {
+                let mut reporter = crate::CliProgressReporter::new();
+                ollama_client
+                    .pull_with_reporter(model, &mut reporter)
+                    .await?;
+            }
+        }
+        Err(err) => {
+            // Not fatal; higher layers may still proceed and surface errors later.
+            tracing::warn!("Failed to query local models from Ollama: {}.", err);
+        }
+    }
+
+    Ok(())
+}

codex-rs/ollama/src/parser.rs

@@ -0,0 +1,82 @@
+use serde_json::Value as JsonValue;
+
+use crate::pull::PullEvent;
+
+// Convert a single JSON object representing a pull update into one or more events.
+pub(crate) fn pull_events_from_value(value: &JsonValue) -> Vec<PullEvent> {
+    let mut events = Vec::new();
+    if let Some(status) = value.get("status").and_then(|s| s.as_str()) {
+        events.push(PullEvent::Status(status.to_string()));
+        if status == "success" {
+            events.push(PullEvent::Success);
+        }
+    }
+    let digest = value
+        .get("digest")
+        .and_then(|d| d.as_str())
+        .unwrap_or("")
+        .to_string();
+    let total = value.get("total").and_then(|t| t.as_u64());
+    let completed = value.get("completed").and_then(|t| t.as_u64());
+    if total.is_some() || completed.is_some() {
+        events.push(PullEvent::ChunkProgress {
+            digest,
+            total,
+            completed,
+        });
+    }
+    events
+}
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+
+    #[test]
+    fn test_pull_events_decoder_status_and_success() {
+        let v: JsonValue = serde_json::json!({"status":"verifying"});
+        let events = pull_events_from_value(&v);
+        assert!(matches!(events.as_slice(), [PullEvent::Status(s)] if s == "verifying"));
+
+        let v2: JsonValue = serde_json::json!({"status":"success"});
+        let events2 = pull_events_from_value(&v2);
+        assert_eq!(events2.len(), 2);
+        assert!(matches!(events2[0], PullEvent::Status(ref s) if s == "success"));
+        assert!(matches!(events2[1], PullEvent::Success));
+    }
+
+    #[test]
+    fn test_pull_events_decoder_progress() {
+        let v: JsonValue = serde_json::json!({"digest":"sha256:abc","total":100});
+        let events = pull_events_from_value(&v);
+        assert_eq!(events.len(), 1);
+        match &events[0] {
+            PullEvent::ChunkProgress {
+                digest,
+                total,
+                completed,
+            } => {
+                assert_eq!(digest, "sha256:abc");
+                assert_eq!(*total, Some(100));
+                assert_eq!(*completed, None);
+            }
+            _ => panic!("expected ChunkProgress"),
+        }
+
+        let v2: JsonValue = serde_json::json!({"digest":"sha256:def","completed":42});
+        let events2 = pull_events_from_value(&v2);
+        assert_eq!(events2.len(), 1);
+        match &events2[0] {
+            PullEvent::ChunkProgress {
+                digest,
+                total,
+                completed,
+            } => {
+                assert_eq!(digest, "sha256:def");
+                assert_eq!(*total, None);
+                assert_eq!(*completed, Some(42));
+            }
+            _ => panic!("expected ChunkProgress"),
+        }
+    }
+}

codex-rs/ollama/src/pull.rs

@@ -0,0 +1,147 @@
+use std::collections::HashMap;
+use std::io;
+use std::io::Write;
+
+/// Events emitted while pulling a model from Ollama.
+#[derive(Debug, Clone)]
+pub enum PullEvent {
+    /// A human-readable status message (e.g., "verifying", "writing").
+    Status(String),
+    /// Byte-level progress update for a specific layer digest.
+    ChunkProgress {
+        digest: String,
+        total: Option<u64>,
+        completed: Option<u64>,
+    },
+    /// The pull finished successfully.
+    Success,
+
+    /// Error event with a message.
+    Error(String),
+}
+
+/// A simple observer for pull progress events. Implementations decide how to
+/// render progress (CLI, TUI, logs, ...).
+pub trait PullProgressReporter {
+    fn on_event(&mut self, event: &PullEvent) -> io::Result<()>;
+}
+
+/// A minimal CLI reporter that writes inline progress to stderr.
+pub struct CliProgressReporter {
+    printed_header: bool,
+    last_line_len: usize,
+    last_completed_sum: u64,
+    last_instant: std::time::Instant,
+    totals_by_digest: HashMap<String, (u64, u64)>,
+}
+
+impl Default for CliProgressReporter {
+    fn default() -> Self {
+        Self::new()
+    }
+}
+
+impl CliProgressReporter {
+    pub fn new() -> Self {
+        Self {
+            printed_header: false,
+            last_line_len: 0,
+            last_completed_sum: 0,
+            last_instant: std::time::Instant::now(),
+            totals_by_digest: HashMap::new(),
+        }
+    }
+}
+
+impl PullProgressReporter for CliProgressReporter {
+    fn on_event(&mut self, event: &PullEvent) -> io::Result<()> {
+        let mut out = std::io::stderr();
+        match event {
+            PullEvent::Status(status) => {
+                // Avoid noisy manifest messages; otherwise show status inline.
+                if status.eq_ignore_ascii_case("pulling manifest") {
+                    return Ok(());
+                }
+                let pad = self.last_line_len.saturating_sub(status.len());
+                let line = format!("\r{status}{}", " ".repeat(pad));
+                self.last_line_len = status.len();
+                out.write_all(line.as_bytes())?;
+                out.flush()
+            }
+            PullEvent::ChunkProgress {
+                digest,
+                total,
+                completed,
+            } => {
+                if let Some(t) = *total {
+                    self.totals_by_digest
+                        .entry(digest.clone())
+                        .or_insert((0, 0))
+                        .0 = t;
+                }
+                if let Some(c) = *completed {
+                    self.totals_by_digest
+                        .entry(digest.clone())
+                        .or_insert((0, 0))
+                        .1 = c;
+                }
+
+                let (sum_total, sum_completed) = self
+                    .totals_by_digest
+                    .values()
+                    .fold((0u64, 0u64), |acc, (t, c)| (acc.0 + *t, acc.1 + *c));
+                if sum_total > 0 {
+                    if !self.printed_header {
+                        let gb = (sum_total as f64) / (1024.0 * 1024.0 * 1024.0);
+                        let header = format!("Downloading model: total {gb:.2} GB\n");
+                        out.write_all(b"\r\x1b[2K")?;
+                        out.write_all(header.as_bytes())?;
+                        self.printed_header = true;
+                    }
+                    let now = std::time::Instant::now();
+                    let dt = now
+                        .duration_since(self.last_instant)
+                        .as_secs_f64()
+                        .max(0.001);
+                    let dbytes = sum_completed.saturating_sub(self.last_completed_sum) as f64;
+                    let speed_mb_s = dbytes / (1024.0 * 1024.0) / dt;
+                    self.last_completed_sum = sum_completed;
+                    self.last_instant = now;
+
+                    let done_gb = (sum_completed as f64) / (1024.0 * 1024.0 * 1024.0);
+                    let total_gb = (sum_total as f64) / (1024.0 * 1024.0 * 1024.0);
+                    let pct = (sum_completed as f64) * 100.0 / (sum_total as f64);
+                    let text =
+                        format!("{done_gb:.2}/{total_gb:.2} GB ({pct:.1}%) {speed_mb_s:.1} MB/s");
+                    let pad = self.last_line_len.saturating_sub(text.len());
+                    let line = format!("\r{text}{}", " ".repeat(pad));
+                    self.last_line_len = text.len();
+                    out.write_all(line.as_bytes())?;
+                    out.flush()
+                } else {
+                    Ok(())
+                }
+            }
+            PullEvent::Error(_) => {
+                // This will be handled by the caller, so we don't do anything
+                // here or the error will be printed twice.
+                Ok(())
+            }
+            PullEvent::Success => {
+                out.write_all(b"\n")?;
+                out.flush()
+            }
+        }
+    }
+}
+
+/// For now the TUI reporter delegates to the CLI reporter. This keeps UI and
+/// CLI behavior aligned until a dedicated TUI integration is implemented.
+#[derive(Default)]
+pub struct TuiProgressReporter(CliProgressReporter);
+
+impl PullProgressReporter for TuiProgressReporter {
+    fn on_event(&mut self, event: &PullEvent) -> io::Result<()> {
+        self.0.on_event(event)
+    }
+}

codex-rs/ollama/src/url.rs

@@ -0,0 +1,39 @@
+/// Identify whether a base_url points at an OpenAI-compatible root (".../v1").
+pub(crate) fn is_openai_compatible_base_url(base_url: &str) -> bool {
+    base_url.trim_end_matches('/').ends_with("/v1")
+}
+
+/// Convert a provider base_url into the native Ollama host root.
+/// For example, "http://localhost:11434/v1" -> "http://localhost:11434".
+pub fn base_url_to_host_root(base_url: &str) -> String {
+    let trimmed = base_url.trim_end_matches('/');
+    if trimmed.ends_with("/v1") {
+        trimmed
+            .trim_end_matches("/v1")
+            .trim_end_matches('/')
+            .to_string()
+    } else {
+        trimmed.to_string()
+    }
+}
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+
+    #[test]
+    fn test_base_url_to_host_root() {
+        assert_eq!(
+            base_url_to_host_root("http://localhost:11434/v1"),
+            "http://localhost:11434"
+        );
+        assert_eq!(
+            base_url_to_host_root("http://localhost:11434"),
+            "http://localhost:11434"
+        );
+        assert_eq!(
+            base_url_to_host_root("http://localhost:11434/"),
+            "http://localhost:11434"
+        );
+    }
+}

codex-rs/scripts/create_github_release.sh

@@ -19,7 +19,33 @@ if ! git diff --quiet || ! git diff --cached --quiet || [ -n "$(git ls-files --o
 fi
 
 # Fail if in a detached HEAD state.
-CURRENT_BRANCH=$(git symbolic-ref --short -q HEAD)
+CURRENT_BRANCH=$(git symbolic-ref --short -q HEAD 2>/dev/null || true)
+if [ -z "${CURRENT_BRANCH:-}" ]; then
+  echo "ERROR: Could not determine the current branch (detached HEAD?)." >&2
+  echo "       Please run this script from a checked-out branch." >&2
+  exit 1
+fi
+
+# Ensure we are on the 'main' branch before proceeding.
+if [ "${CURRENT_BRANCH}" != "main" ]; then
+  echo "ERROR: Releases must be created from the 'main' branch (current: '${CURRENT_BRANCH}')." >&2
+  echo "       Please switch to 'main' and try again." >&2
+  exit 1
+fi
+
+# Ensure the current local commit on 'main' is present on 'origin/main'.
+# This guarantees we only create releases from commits that are already on
+# the canonical repository (https://github.com/openai/codex).
+if ! git fetch --quiet origin main; then
+  echo "ERROR: Failed to fetch 'origin/main'. Ensure the 'origin' remote is configured and reachable." >&2
+  exit 1
+fi
+
+if ! git merge-base --is-ancestor HEAD origin/main; then
+  echo "ERROR: Your local 'main' HEAD commit is not present on 'origin/main'." >&2
+  echo "       Please push your commits first (git push origin main) or check out a commit on 'origin/main'." >&2
+  exit 1
+fi
 
 # Create a new branch for the release and make a commit with the new version.
 if [ $# -ge 1 ]; then

codex-rs/tui/Cargo.toml

@@ -33,6 +33,7 @@ codex-common = { path = "../common", features = [
 codex-core = { path = "../core" }
 codex-file-search = { path = "../file-search" }
 codex-login = { path = "../login" }
+codex-ollama = { path = "../ollama" }
 color-eyre = "0.6.3"
 crossterm = { version = "0.28.1", features = ["bracketed-paste"] }
 image = { version = "^0.25.6", default-features = false, features = ["jpeg"] }

codex-rs/tui/prompt_for_init_command.md

@@ -0,0 +1,40 @@
+Generate a file named AGENTS.md that serves as a contributor guide for this repository.
+Your goal is to produce a clear, concise, and well-structured document with descriptive headings and actionable explanations for each section.
+Follow the outline below, but adapt as needed — add sections if relevant, and omit those that do not apply to this project.
+
+Document Requirements
+
+- Title the document "Repository Guidelines".
+- Use Markdown headings (#, ##, etc.) for structure.
+- Keep the document concise. 200-400 words is optimal.
+- Keep explanations short, direct, and specific to this repository.
+- Provide examples where helpful (commands, directory paths, naming patterns).
+- Maintain a professional, instructional tone.
+
+Recommended Sections
+
+Project Structure & Module Organization
+
+- Outline the project structure, including where the source code, tests, and assets are located.
+
+Build, Test, and Development Commands
+
+- List key commands for building, testing, and running locally (e.g., npm test, make build).
+- Briefly explain what each command does.
+
+Coding Style & Naming Conventions
+
+- Specify indentation rules, language-specific style preferences, and naming patterns.
+- Include any formatting or linting tools used.
+
+Testing Guidelines
+
+- Identify testing frameworks and coverage requirements.
+- State test naming conventions and how to run tests.
+
+Commit & Pull Request Guidelines
+
+- Summarize commit message conventions found in the project’s Git history.
+- Outline pull request requirements (descriptions, linked issues, screenshots, etc.).
+
+(Optional) Add other sections if relevant, such as Security & Configuration Tips, Architecture Overview, or Agent-Specific Instructions.

codex-rs/tui/src/app.rs

@@ -5,6 +5,10 @@ use crate::file_search::FileSearchManager;
 use crate::get_git_diff::get_git_diff;
 use crate::git_warning_screen::GitWarningOutcome;
 use crate::git_warning_screen::GitWarningScreen;
+use crate::onboarding::onboarding_screen::KeyEventResult;
+use crate::onboarding::onboarding_screen::KeyboardHandler;
+use crate::onboarding::onboarding_screen::OnboardingScreen;
+use crate::should_show_login_screen;
 use crate::slash_command::SlashCommand;
 use crate::tui;
 use codex_core::config::Config;
@@ -35,6 +39,9 @@ const REDRAW_DEBOUNCE: Duration = Duration::from_millis(10);
 /// Top-level application state: which full-screen view is currently active.
 #[allow(clippy::large_enum_variant)]
 enum AppState<'a> {
+    Onboarding {
+        screen: OnboardingScreen,
+    },
     /// The main chat UI is visible.
     Chat {
         /// Boxed to avoid a large enum variant and reduce the overall size of
@@ -42,7 +49,9 @@ enum AppState<'a> {
         widget: Box<ChatWidget<'a>>,
     },
     /// The start-up warning that recommends running codex inside a Git repo.
-    GitWarning { screen: GitWarningScreen },
+    GitWarning {
+        screen: GitWarningScreen,
+    },
 }
 
 pub(crate) struct App<'a> {
@@ -133,7 +142,20 @@ impl App<'_> {
             });
         }
 
-        let (app_state, chat_args) = if show_git_warning {
+        let show_login_screen = should_show_login_screen(&config);
+        let (app_state, chat_args) = if show_login_screen {
+            (
+                AppState::Onboarding {
+                    screen: OnboardingScreen::new(app_event_tx.clone(), config.codex_home.clone()),
+                },
+                Some(ChatWidgetArgs {
+                    config: config.clone(),
+                    initial_prompt,
+                    initial_images,
+                    enhanced_keys_supported,
+                }),
+            )
+        } else if show_git_warning {
             (
                 AppState::GitWarning {
                     screen: GitWarningScreen::new(),
@@ -232,9 +254,23 @@ impl App<'_> {
                                 AppState::Chat { widget } => {
                                     widget.on_ctrl_c();
                                 }
-                                AppState::GitWarning { .. } => {
-                                    // No-op.
+                                AppState::Onboarding { .. } => {
+                                    self.app_event_tx.send(AppEvent::ExitRequest);
                                 }
+                                AppState::GitWarning { .. } => {
+                                    // Allow exiting the app with Ctrl+C from the warning screen.
+                                    self.app_event_tx.send(AppEvent::ExitRequest);
+                                }
+                            }
+                        }
+                        KeyEvent {
+                            code: KeyCode::Char('z'),
+                            modifiers: crossterm::event::KeyModifiers::CONTROL,
+                            kind: KeyEventKind::Press,
+                            ..
+                        } => {
+                            if let AppState::Chat { widget } = &mut self.app_state {
+                                widget.on_ctrl_z();
                             }
                         }
                         KeyEvent {
@@ -254,6 +290,9 @@ impl App<'_> {
                                         self.dispatch_key_event(key_event);
                                     }
                                 }
+                                AppState::Onboarding { .. } => {
+                                    self.app_event_tx.send(AppEvent::ExitRequest);
+                                }
                                 AppState::GitWarning { .. } => {
                                     self.app_event_tx.send(AppEvent::ExitRequest);
                                 }
@@ -281,10 +320,12 @@ impl App<'_> {
                 }
                 AppEvent::CodexOp(op) => match &mut self.app_state {
                     AppState::Chat { widget } => widget.submit_op(op),
+                    AppState::Onboarding { .. } => {}
                     AppState::GitWarning { .. } => {}
                 },
                 AppEvent::LatestLog(line) => match &mut self.app_state {
                     AppState::Chat { widget } => widget.update_latest_log(line),
+                    AppState::Onboarding { .. } => {}
                     AppState::GitWarning { .. } => {}
                 },
                 AppEvent::DispatchCommand(command) => match command {
@@ -299,6 +340,13 @@ impl App<'_> {
                         self.app_state = AppState::Chat { widget: new_widget };
                         self.app_event_tx.send(AppEvent::RequestRedraw);
                     }
+                    SlashCommand::Init => {
+                        // Guard: do not run if a task is active.
+                        if let AppState::Chat { widget } = &mut self.app_state {
+                            const INIT_PROMPT: &str = include_str!("../prompt_for_init_command.md");
+                            widget.submit_text_message(INIT_PROMPT.to_string());
+                        }
+                    }
                     SlashCommand::Compact => {
                         if let AppState::Chat { widget } = &mut self.app_state {
                             widget.clear_token_usage();
@@ -329,6 +377,11 @@ impl App<'_> {
                             widget.add_diff_output(text);
                         }
                     }
+                    SlashCommand::Status => {
+                        if let AppState::Chat { widget } = &mut self.app_state {
+                            widget.add_status_output();
+                        }
+                    }
                     #[cfg(debug_assertions)]
                     SlashCommand::TestApproval => {
                         use std::collections::HashMap;
@@ -359,6 +412,7 @@ impl App<'_> {
                                             FileChange::Update {
                                                 unified_diff: "+test\n-test2".to_string(),
                                                 move_path: None,
+                                                hunks: None,
                                             },
                                         ),
                                     ]),
@@ -369,6 +423,12 @@ impl App<'_> {
                         }));
                     }
                 },
+                AppEvent::OnboardingAuthComplete(result) => {
+                    if let AppState::Onboarding { screen } = &mut self.app_state {
+                        // Let the onboarding screen handle success/failure and emit follow-up events.
+                        let _ = screen.on_auth_complete(result);
+                    }
+                }
                 AppEvent::StartFileSearch(query) => {
                     self.file_search.on_user_query(query);
                 }
@@ -387,6 +447,7 @@ impl App<'_> {
     pub(crate) fn token_usage(&self) -> codex_core::protocol::TokenUsage {
         match &self.app_state {
             AppState::Chat { widget } => widget.token_usage().clone(),
+            AppState::Onboarding { .. } => codex_core::protocol::TokenUsage::default(),
             AppState::GitWarning { .. } => codex_core::protocol::TokenUsage::default(),
         }
     }
@@ -415,7 +476,8 @@ impl App<'_> {
         let size = terminal.size()?;
         let desired_height = match &self.app_state {
             AppState::Chat { widget } => widget.desired_height(size.width),
-            AppState::GitWarning { .. } => 10,
+            AppState::Onboarding { .. } => size.height,
+            AppState::GitWarning { .. } => size.height,
         };
 
         let mut area = terminal.viewport_area;
@@ -445,6 +507,7 @@ impl App<'_> {
                 }
                 frame.render_widget_ref(&**widget, frame.area())
             }
+            AppState::Onboarding { screen } => frame.render_widget_ref(&*screen, frame.area()),
             AppState::GitWarning { screen } => frame.render_widget_ref(&*screen, frame.area()),
         })?;
         Ok(())
@@ -457,6 +520,25 @@ impl App<'_> {
             AppState::Chat { widget } => {
                 widget.handle_key_event(key_event);
             }
+            AppState::Onboarding { screen } => match screen.handle_key_event(key_event) {
+                KeyEventResult::Continue => {
+                    self.app_state = AppState::Chat {
+                        widget: Box::new(ChatWidget::new(
+                            self.config.clone(),
+                            self.app_event_tx.clone(),
+                            None,
+                            Vec::new(),
+                            self.enhanced_keys_supported,
+                        )),
+                    };
+                }
+                KeyEventResult::Quit => {
+                    self.app_event_tx.send(AppEvent::ExitRequest);
+                }
+                KeyEventResult::None => {
+                    // do nothing
+                }
+            },
             AppState::GitWarning { screen } => match screen.handle_key_event(key_event) {
                 GitWarningOutcome::Continue => {
                     // User accepted – switch to chat view.
@@ -488,6 +570,7 @@ impl App<'_> {
     fn dispatch_paste_event(&mut self, pasted: String) {
         match &mut self.app_state {
             AppState::Chat { widget } => widget.handle_paste(pasted),
+            AppState::Onboarding { .. } => {}
             AppState::GitWarning { .. } => {}
         }
     }
@@ -495,6 +578,7 @@ impl App<'_> {
     fn dispatch_codex_event(&mut self, event: Event) {
         match &mut self.app_state {
             AppState::Chat { widget } => widget.handle_codex_event(event),
+            AppState::Onboarding { .. } => {}
             AppState::GitWarning { .. } => {}
         }
     }

codex-rs/tui/src/app_event.rs

@@ -48,4 +48,7 @@ pub(crate) enum AppEvent {
     },
 
     InsertHistory(Vec<Line<'static>>),
+
+    /// Onboarding: result of login_with_chatgpt.
+    OnboardingAuthComplete(Result<(), String>),
 }

codex-rs/tui/src/bottom_pane/chat_composer.rs

@@ -729,6 +729,7 @@ impl WidgetRef for &ChatComposer {
 
 #[cfg(test)]
 mod tests {
+    use crate::app_event::AppEvent;
     use crate::bottom_pane::AppEventSender;
     use crate::bottom_pane::ChatComposer;
     use crate::bottom_pane::InputResult;
@@ -1004,6 +1005,49 @@ mod tests {
         }
     }
 
+    #[test]
+    fn slash_init_dispatches_command_and_does_not_submit_literal_text() {
+        use crossterm::event::KeyCode;
+        use crossterm::event::KeyEvent;
+        use crossterm::event::KeyModifiers;
+        use std::sync::mpsc::TryRecvError;
+
+        let (tx, rx) = std::sync::mpsc::channel();
+        let sender = AppEventSender::new(tx);
+        let mut composer = ChatComposer::new(true, sender, false);
+
+        // Type the slash command.
+        for ch in [
+            '/', 'i', 'n', 'i', 't', // "/init"
+        ] {
+            let _ = composer.handle_key_event(KeyEvent::new(KeyCode::Char(ch), KeyModifiers::NONE));
+        }
+
+        // Press Enter to dispatch the selected command.
+        let (result, _needs_redraw) =
+            composer.handle_key_event(KeyEvent::new(KeyCode::Enter, KeyModifiers::NONE));
+
+        // When a slash command is dispatched, the composer should not submit
+        // literal text and should clear its textarea.
+        match result {
+            InputResult::None => {}
+            InputResult::Submitted(text) => {
+                panic!("expected command dispatch, but composer submitted literal text: {text}")
+            }
+        }
+        assert!(composer.textarea.is_empty(), "composer should be cleared");
+
+        // Verify a DispatchCommand event for the "init" command was sent.
+        match rx.try_recv() {
+            Ok(AppEvent::DispatchCommand(cmd)) => {
+                assert_eq!(cmd.command(), "init");
+            }
+            Ok(_other) => panic!("unexpected app event"),
+            Err(TryRecvError::Empty) => panic!("expected a DispatchCommand event for '/init'"),
+            Err(TryRecvError::Disconnected) => panic!("app event channel disconnected"),
+        }
+    }
+
     #[test]
     fn test_multiple_pastes_submission() {
         use crossterm::event::KeyCode;

codex-rs/tui/src/bottom_pane/command_popup.rs

@@ -188,3 +188,38 @@ impl WidgetRef for CommandPopup {
         table.render(area, buf);
     }
 }
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+
+    #[test]
+    fn filter_includes_init_when_typing_prefix() {
+        let mut popup = CommandPopup::new();
+        // Simulate the composer line starting with '/in' so the popup filters
+        // matching commands by prefix.
+        popup.on_composer_text_change("/in".to_string());
+
+        // Access the filtered list via the selected command and ensure that
+        // one of the matches is the new "init" command.
+        let matches = popup.filtered_commands();
+        assert!(
+            matches.iter().any(|cmd| cmd.command() == "init"),
+            "expected '/init' to appear among filtered commands"
+        );
+    }
+
+    #[test]
+    fn selecting_init_by_exact_match() {
+        let mut popup = CommandPopup::new();
+        popup.on_composer_text_change("/init".to_string());
+
+        // When an exact match exists, the selected command should be that
+        // command by default.
+        let selected = popup.selected_command();
+        match selected {
+            Some(cmd) => assert_eq!(cmd.command(), "init"),
+            None => panic!("expected a selected command for exact match"),
+        }
+    }
+}

codex-rs/tui/src/bottom_pane/mod.rs

@@ -138,6 +138,11 @@ impl BottomPane<'_> {
             view.handle_key_event(self, key_event);
             if !view.is_complete() {
                 self.active_view = Some(view);
+            } else if self.is_task_running {
+                let mut v = StatusIndicatorView::new(self.app_event_tx.clone());
+                v.update_text("waiting for model".to_string());
+                self.active_view = Some(Box::new(v));
+                self.status_view_active = true;
             }
             self.request_redraw();
             InputResult::None
@@ -163,6 +168,12 @@ impl BottomPane<'_> {
             CancellationEvent::Handled => {
                 if !view.is_complete() {
                     self.active_view = Some(view);
+                } else if self.is_task_running {
+                    // Modal aborted but task still running – restore status indicator.
+                    let mut v = StatusIndicatorView::new(self.app_event_tx.clone());
+                    v.update_text("waiting for model".to_string());
+                    self.active_view = Some(Box::new(v));
+                    self.status_view_active = true;
                 }
                 self.show_ctrl_c_quit_hint();
             }
@@ -202,15 +213,20 @@ impl BottomPane<'_> {
             handled_by_view = true;
         }
 
-        // Fallback: if the current active view did not consume status updates,
-        // present an overlay above the composer.
-        if !handled_by_view {
+        // Fallback: if the current active view did not consume status updates
+        // and no modal view is active, present an overlay above the composer.
+        // If a modal is active, do NOT render the overlay to avoid drawing
+        // over the dialog.
+        if !handled_by_view && self.active_view.is_none() {
             if self.live_status.is_none() {
                 self.live_status = Some(StatusIndicatorWidget::new(self.app_event_tx.clone()));
             }
             if let Some(status) = &mut self.live_status {
                 status.update_text(text);
             }
+        } else if !handled_by_view {
+            // Ensure any previous overlay is cleared when a modal becomes active.
+            self.live_status = None;
         }
         self.request_redraw();
     }
@@ -296,6 +312,8 @@ impl BottomPane<'_> {
         // Otherwise create a new approval modal overlay.
         let modal = ApprovalModalView::new(request, self.app_event_tx.clone());
         self.active_view = Some(Box::new(modal));
+        // Hide any overlay status while a modal is visible.
+        self.live_status = None;
         self.status_view_active = false;
         self.request_redraw()
     }
@@ -368,16 +386,18 @@ impl WidgetRef for &BottomPane<'_> {
             y_offset = y_offset.saturating_add(1);
         }
         if let Some(status) = &self.live_status {
-            let live_h = status.desired_height(area.width).min(area.height);
+            let live_h = status
+                .desired_height(area.width)
+                .min(area.height.saturating_sub(y_offset));
             if live_h > 0 {
                 let live_rect = Rect {
                     x: area.x,
-                    y: area.y,
+                    y: area.y + y_offset,
                     width: area.width,
                     height: live_h,
                 };
                 status.render_ref(live_rect, buf);
-                y_offset = live_h;
+                y_offset = y_offset.saturating_add(live_h);
             }
         }
 
@@ -540,6 +560,122 @@ mod tests {
         );
     }
 
+    #[test]
+    fn overlay_not_shown_above_approval_modal() {
+        let (tx_raw, _rx) = channel::<AppEvent>();
+        let tx = AppEventSender::new(tx_raw);
+        let mut pane = BottomPane::new(BottomPaneParams {
+            app_event_tx: tx,
+            has_input_focus: true,
+            enhanced_keys_supported: false,
+        });
+
+        // Create an approval modal (active view).
+        pane.push_approval_request(exec_request());
+        // Attempt to update status; this should NOT create an overlay while modal is visible.
+        pane.update_status_text("running command".to_string());
+
+        // Render and verify the top row does not include the Working header overlay.
+        let area = Rect::new(0, 0, 60, 6);
+        let mut buf = Buffer::empty(area);
+        (&pane).render_ref(area, &mut buf);
+
+        let mut r0 = String::new();
+        for x in 0..area.width {
+            r0.push(buf[(x, 0)].symbol().chars().next().unwrap_or(' '));
+        }
+        assert!(
+            !r0.contains("Working"),
+            "overlay Working header should not render above modal"
+        );
+    }
+
+    #[test]
+    fn composer_not_shown_after_denied_if_task_running() {
+        let (tx_raw, rx) = channel::<AppEvent>();
+        let tx = AppEventSender::new(tx_raw);
+        let mut pane = BottomPane::new(BottomPaneParams {
+            app_event_tx: tx.clone(),
+            has_input_focus: true,
+            enhanced_keys_supported: false,
+        });
+
+        // Start a running task so the status indicator replaces the composer.
+        pane.set_task_running(true);
+        pane.update_status_text("waiting for model".to_string());
+
+        // Push an approval modal (e.g., command approval) which should hide the status view.
+        pane.push_approval_request(exec_request());
+
+        // Simulate pressing 'n' (deny) on the modal.
+        use crossterm::event::KeyCode;
+        use crossterm::event::KeyEvent;
+        use crossterm::event::KeyModifiers;
+        pane.handle_key_event(KeyEvent::new(KeyCode::Char('n'), KeyModifiers::NONE));
+
+        // After denial, since the task is still running, the status indicator
+        // should be restored as the active view; the composer should NOT be visible.
+        assert!(
+            pane.status_view_active,
+            "status view should be active after denial"
+        );
+        assert!(pane.active_view.is_some(), "active view should be present");
+
+        // Render and ensure the top row includes the Working header instead of the composer.
+        // Give the animation thread a moment to tick.
+        std::thread::sleep(std::time::Duration::from_millis(120));
+        let area = Rect::new(0, 0, 40, 3);
+        let mut buf = Buffer::empty(area);
+        (&pane).render_ref(area, &mut buf);
+        let mut row0 = String::new();
+        for x in 0..area.width {
+            row0.push(buf[(x, 0)].symbol().chars().next().unwrap_or(' '));
+        }
+        assert!(
+            row0.contains("Working"),
+            "expected Working header after denial: {row0:?}"
+        );
+
+        // Drain the channel to avoid unused warnings.
+        drop(rx);
+    }
+
+    #[test]
+    fn status_indicator_visible_during_command_execution() {
+        let (tx_raw, _rx) = channel::<AppEvent>();
+        let tx = AppEventSender::new(tx_raw);
+        let mut pane = BottomPane::new(BottomPaneParams {
+            app_event_tx: tx,
+            has_input_focus: true,
+            enhanced_keys_supported: false,
+        });
+
+        // Begin a task: show initial status.
+        pane.set_task_running(true);
+        pane.update_status_text("waiting for model".to_string());
+
+        // As a long-running command begins (post-approval), ensure the status
+        // indicator is visible while we wait for the command to run.
+        pane.update_status_text("running command".to_string());
+
+        // Allow some frames so the animation thread ticks.
+        std::thread::sleep(std::time::Duration::from_millis(120));
+
+        // Render and confirm the line contains the "Working" header.
+        let area = Rect::new(0, 0, 40, 3);
+        let mut buf = Buffer::empty(area);
+        (&pane).render_ref(area, &mut buf);
+
+        let mut row0 = String::new();
+        for x in 0..area.width {
+            row0.push(buf[(x, 0)].symbol().chars().next().unwrap_or(' '));
+        }
+        assert!(
+            row0.contains("Working"),
+            "expected Working header: {row0:?}"
+        );
+    }
+
     #[test]
     fn bottom_padding_present_for_status_view() {
         let (tx_raw, _rx) = channel::<AppEvent>();

codex-rs/tui/src/bottom_pane/textarea.rs

@@ -206,7 +206,10 @@ impl TextArea {
         match event {
             KeyEvent {
                 code: KeyCode::Char(c),
-                modifiers: KeyModifiers::NONE | KeyModifiers::SHIFT | KeyModifiers::ALT,
+                // Insert plain characters (and Shift-modified). Do NOT insert when ALT is held,
+                // because many terminals map Option/Meta combos to ALT+<char> (e.g. ESC f/ESC b)
+                // for word navigation. Those are handled explicitly below.
+                modifiers: KeyModifiers::NONE | KeyModifiers::SHIFT,
                 ..
             } => self.insert_str(&c.to_string()),
             KeyEvent {
@@ -245,6 +248,23 @@ impl TextArea {
             } => {
                 self.delete_backward_word();
             }
+            // Meta-b -> move to beginning of previous word
+            // Meta-f -> move to end of next word
+            // Many terminals map Option (macOS) to Alt. Some send Alt|Shift, so match contains(ALT).
+            KeyEvent {
+                code: KeyCode::Char('b'),
+                modifiers: KeyModifiers::ALT,
+                ..
+            } => {
+                self.set_cursor(self.beginning_of_previous_word());
+            }
+            KeyEvent {
+                code: KeyCode::Char('f'),
+                modifiers: KeyModifiers::ALT,
+                ..
+            } => {
+                self.set_cursor(self.end_of_next_word());
+            }
             KeyEvent {
                 code: KeyCode::Char('u'),
                 modifiers: KeyModifiers::CONTROL,
@@ -275,6 +295,23 @@ impl TextArea {
             } => {
                 self.move_cursor_right();
             }
+            // Some terminals send Alt+Arrow for word-wise movement:
+            // Option/Left -> Alt+Left (previous word start)
+            // Option/Right -> Alt+Right (next word end)
+            KeyEvent {
+                code: KeyCode::Left,
+                modifiers: KeyModifiers::ALT,
+                ..
+            } => {
+                self.set_cursor(self.beginning_of_previous_word());
+            }
+            KeyEvent {
+                code: KeyCode::Right,
+                modifiers: KeyModifiers::ALT,
+                ..
+            } => {
+                self.set_cursor(self.end_of_next_word());
+            }
             KeyEvent {
                 code: KeyCode::Up, ..
             } => {
@@ -312,20 +349,6 @@ impl TextArea {
             } => {
                 self.move_cursor_to_end_of_line(true);
             }
-            KeyEvent {
-                code: KeyCode::Left,
-                modifiers: KeyModifiers::CONTROL | KeyModifiers::ALT,
-                ..
-            } => {
-                self.set_cursor(self.beginning_of_previous_word());
-            }
-            KeyEvent {
-                code: KeyCode::Right,
-                modifiers: KeyModifiers::CONTROL | KeyModifiers::ALT,
-                ..
-            } => {
-                self.set_cursor(self.end_of_next_word());
-            }
             o => {
                 tracing::debug!("Unhandled key event in TextArea: {:?}", o);
             }

codex-rs/tui/src/chatwidget.rs

@@ -9,7 +9,10 @@ use codex_core::protocol::AgentMessageDeltaEvent;
 use codex_core::protocol::AgentMessageEvent;
 use codex_core::protocol::AgentReasoningDeltaEvent;
 use codex_core::protocol::AgentReasoningEvent;
+use codex_core::protocol::AgentReasoningRawContentDeltaEvent;
+use codex_core::protocol::AgentReasoningRawContentEvent;
 use codex_core::protocol::ApplyPatchApprovalRequestEvent;
+use codex_core::protocol::BackgroundEventEvent;
 use codex_core::protocol::ErrorEvent;
 use codex_core::protocol::Event;
 use codex_core::protocol::EventMsg;
@@ -23,14 +26,18 @@ use codex_core::protocol::Op;
 use codex_core::protocol::PatchApplyBeginEvent;
 use codex_core::protocol::TaskCompleteEvent;
 use codex_core::protocol::TokenUsage;
+use codex_core::protocol::TurnDiffEvent;
 use crossterm::event::KeyEvent;
 use crossterm::event::KeyEventKind;
 use ratatui::buffer::Buffer;
+use ratatui::layout::Constraint;
+use ratatui::layout::Layout;
 use ratatui::layout::Rect;
 use ratatui::widgets::Widget;
 use ratatui::widgets::WidgetRef;
 use tokio::sync::mpsc::UnboundedSender;
 use tokio::sync::mpsc::unbounded_channel;
+use tracing::info;
 
 use crate::app_event::AppEvent;
 use crate::app_event_sender::AppEventSender;
@@ -57,10 +64,12 @@ pub(crate) struct ChatWidget<'a> {
     app_event_tx: AppEventSender,
     codex_op_tx: UnboundedSender<Op>,
     bottom_pane: BottomPane<'a>,
+    active_history_cell: Option<HistoryCell>,
     config: Config,
     initial_user_message: Option<UserMessage>,
     token_usage: TokenUsage,
     reasoning_buffer: String,
+    content_buffer: String,
     // Buffer for streaming assistant answer text; we do not surface partial
     // We wait for the final AgentMessage event and then emit the full text
     // at once into scrollback so the history contains a single message.
@@ -101,6 +110,51 @@ fn create_initial_user_message(text: String, image_paths: Vec<PathBuf>) -> Optio
 }
 
 impl ChatWidget<'_> {
+    fn interrupt_running_task(&mut self) {
+        if self.bottom_pane.is_task_running() {
+            self.active_history_cell = None;
+            self.bottom_pane.clear_ctrl_c_quit_hint();
+            self.submit_op(Op::Interrupt);
+            self.bottom_pane.set_task_running(false);
+            self.bottom_pane.clear_live_ring();
+            self.live_builder = RowBuilder::new(self.live_builder.width());
+            self.current_stream = None;
+            self.stream_header_emitted = false;
+            self.answer_buffer.clear();
+            self.reasoning_buffer.clear();
+            self.content_buffer.clear();
+            self.request_redraw();
+        }
+    }
+    fn layout_areas(&self, area: Rect) -> [Rect; 2] {
+        Layout::vertical([
+            Constraint::Max(
+                self.active_history_cell
+                    .as_ref()
+                    .map_or(0, |c| c.desired_height(area.width)),
+            ),
+            Constraint::Min(self.bottom_pane.desired_height(area.width)),
+        ])
+        .areas(area)
+    }
+    fn emit_stream_header(&mut self, kind: StreamKind) {
+        use ratatui::text::Line as RLine;
+        if self.stream_header_emitted {
+            return;
+        }
+        let header = match kind {
+            StreamKind::Reasoning => RLine::from("thinking".magenta().italic()),
+            StreamKind::Answer => RLine::from("codex".magenta().bold()),
+        };
+        self.app_event_tx
+            .send(AppEvent::InsertHistory(vec![header]));
+        self.stream_header_emitted = true;
+    }
+    fn finalize_active_stream(&mut self) {
+        if let Some(kind) = self.current_stream {
+            self.finalize_stream(kind);
+        }
+    }
     pub(crate) fn new(
         config: Config,
         app_event_tx: AppEventSender,
@@ -154,6 +208,7 @@ impl ChatWidget<'_> {
                 has_input_focus: true,
                 enhanced_keys_supported,
             }),
+            active_history_cell: None,
             config,
             initial_user_message: create_initial_user_message(
                 initial_prompt.unwrap_or_default(),
@@ -161,6 +216,7 @@ impl ChatWidget<'_> {
             ),
             token_usage: TokenUsage::default(),
             reasoning_buffer: String::new(),
+            content_buffer: String::new(),
             answer_buffer: String::new(),
             running_commands: HashMap::new(),
             live_builder: RowBuilder::new(80),
@@ -172,6 +228,10 @@ impl ChatWidget<'_> {
 
     pub fn desired_height(&self, width: u16) -> u16 {
         self.bottom_pane.desired_height(width)
+            + self
+                .active_history_cell
+                .as_ref()
+                .map_or(0, |c| c.desired_height(width))
     }
 
     pub(crate) fn handle_key_event(&mut self, key_event: KeyEvent) {
@@ -276,6 +336,20 @@ impl ChatWidget<'_> {
                 self.finalize_stream(StreamKind::Reasoning);
                 self.request_redraw();
             }
+            EventMsg::AgentReasoningRawContentDelta(AgentReasoningRawContentDeltaEvent {
+                delta,
+            }) => {
+                // Treat raw reasoning content the same as summarized reasoning for UI flow.
+                self.begin_stream(StreamKind::Reasoning);
+                self.reasoning_buffer.push_str(&delta);
+                self.stream_push_and_maybe_commit(&delta);
+                self.request_redraw();
+            }
+            EventMsg::AgentReasoningRawContent(AgentReasoningRawContentEvent { text: _ }) => {
+                // Finalize the raw reasoning stream just like the summarized reasoning event.
+                self.finalize_stream(StreamKind::Reasoning);
+                self.request_redraw();
+            }
             EventMsg::TaskStarted => {
                 self.bottom_pane.clear_ctrl_c_quit_hint();
                 self.bottom_pane.set_task_running(true);
@@ -299,6 +373,14 @@ impl ChatWidget<'_> {
             EventMsg::Error(ErrorEvent { message }) => {
                 self.add_to_history(HistoryCell::new_error_event(message.clone()));
                 self.bottom_pane.set_task_running(false);
+                self.bottom_pane.clear_live_ring();
+                self.live_builder = RowBuilder::new(self.live_builder.width());
+                self.current_stream = None;
+                self.stream_header_emitted = false;
+                self.answer_buffer.clear();
+                self.reasoning_buffer.clear();
+                self.content_buffer.clear();
+                self.request_redraw();
             }
             EventMsg::PlanUpdate(update) => {
                 // Commit plan updates directly to history (no status-line preview).
@@ -310,6 +392,7 @@ impl ChatWidget<'_> {
                 cwd,
                 reason,
             }) => {
+                self.finalize_active_stream();
                 // Log a background summary immediately so the history is chronological.
                 let cmdline = strip_bash_lc_and_escape(&command);
                 let text = format!(
@@ -336,6 +419,7 @@ impl ChatWidget<'_> {
                 reason,
                 grant_root,
             }) => {
+                self.finalize_active_stream();
                 // ------------------------------------------------------------------
                 // Before we even prompt the user for approval we surface the patch
                 // summary in the main conversation so that the dialog appears in a
@@ -365,6 +449,10 @@ impl ChatWidget<'_> {
                 command,
                 cwd,
             }) => {
+                self.finalize_active_stream();
+                // Ensure the status indicator is visible while the command runs.
+                self.bottom_pane
+                    .update_status_text("running command".to_string());
                 self.running_commands.insert(
                     call_id,
                     RunningCommand {
@@ -372,9 +460,11 @@ impl ChatWidget<'_> {
                         cwd: cwd.clone(),
                     },
                 );
-                self.add_to_history(HistoryCell::new_active_exec_command(command));
+                self.active_history_cell = Some(HistoryCell::new_active_exec_command(command));
+            }
+            EventMsg::ExecCommandOutputDelta(_) => {
+                // TODO
             }
-            EventMsg::ExecCommandOutputDelta(_) => {}
             EventMsg::PatchApplyBegin(PatchApplyBeginEvent {
                 call_id: _,
                 auto_approved,
@@ -385,6 +475,13 @@ impl ChatWidget<'_> {
                     changes,
                 ));
             }
+            EventMsg::PatchApplyEnd(event) => {
+                self.add_to_history(HistoryCell::new_patch_apply_end(
+                    event.stdout,
+                    event.stderr,
+                    event.success,
+                ));
+            }
             EventMsg::ExecCommandEnd(ExecCommandEndEvent {
                 call_id,
                 exit_code,
@@ -394,6 +491,7 @@ impl ChatWidget<'_> {
             }) => {
                 // Compute summary before moving stdout into the history cell.
                 let cmd = self.running_commands.remove(&call_id);
+                self.active_history_cell = None;
                 self.add_to_history(HistoryCell::new_completed_exec_command(
                     cmd.map(|cmd| cmd.command).unwrap_or_else(|| vec![call_id]),
                     CommandOutput {
@@ -408,6 +506,7 @@ impl ChatWidget<'_> {
                 call_id: _,
                 invocation,
             }) => {
+                self.finalize_active_stream();
                 self.add_to_history(HistoryCell::new_active_mcp_tool_call(invocation));
             }
             EventMsg::McpToolCallEnd(McpToolCallEndEvent {
@@ -441,17 +540,20 @@ impl ChatWidget<'_> {
             EventMsg::ShutdownComplete => {
                 self.app_event_tx.send(AppEvent::ExitRequest);
             }
-            event => {
-                let text = format!("{event:?}");
-                self.add_to_history(HistoryCell::new_background_event(text.clone()));
-                self.update_latest_log(text);
+            EventMsg::TurnDiff(TurnDiffEvent { unified_diff }) => {
+                info!("TurnDiffEvent: {unified_diff}");
+            }
+            EventMsg::BackgroundEvent(BackgroundEventEvent { message }) => {
+                info!("BackgroundEvent: {message}");
             }
         }
     }
 
     /// Update the live log preview while a task is running.
     pub(crate) fn update_latest_log(&mut self, line: String) {
-        self.bottom_pane.update_status_text(line);
+        if self.bottom_pane.is_task_running() {
+            self.bottom_pane.update_status_text(line);
+        }
     }
 
     fn request_redraw(&mut self) {
@@ -462,6 +564,13 @@ impl ChatWidget<'_> {
         self.add_to_history(HistoryCell::new_diff_output(diff_output.clone()));
     }
 
+    pub(crate) fn add_status_output(&mut self) {
+        self.add_to_history(HistoryCell::new_status_output(
+            &self.config,
+            &self.token_usage,
+        ));
+    }
+
     /// Forward file-search results to the bottom pane.
     pub(crate) fn apply_file_search_result(&mut self, query: String, matches: Vec<FileMatch>) {
         self.bottom_pane.on_file_search_result(query, matches);
@@ -476,10 +585,7 @@ impl ChatWidget<'_> {
             CancellationEvent::Ignored => {}
         }
         if self.bottom_pane.is_task_running() {
-            self.bottom_pane.clear_ctrl_c_quit_hint();
-            self.submit_op(Op::Interrupt);
-            self.answer_buffer.clear();
-            self.reasoning_buffer.clear();
+            self.interrupt_running_task();
             CancellationEvent::Ignored
         } else if self.bottom_pane.ctrl_c_quit_hint_visible() {
             self.submit_op(Op::Shutdown);
@@ -490,6 +596,10 @@ impl ChatWidget<'_> {
         }
     }
 
+    pub(crate) fn on_ctrl_z(&mut self) {
+        self.interrupt_running_task();
+    }
+
     pub(crate) fn composer_is_empty(&self) -> bool {
         self.bottom_pane.composer_is_empty()
     }
@@ -501,6 +611,16 @@ impl ChatWidget<'_> {
         }
     }
 
+    /// Programmatically submit a user text message as if typed in the
+    /// composer. The text will be added to conversation history and sent to
+    /// the agent.
+    pub(crate) fn submit_text_message(&mut self, text: String) {
+        if text.is_empty() {
+            return;
+        }
+        self.submit_user_message(text.into());
+    }
+
     pub(crate) fn token_usage(&self) -> &TokenUsage {
         &self.token_usage
     }
@@ -512,12 +632,19 @@ impl ChatWidget<'_> {
     }
 
     pub fn cursor_pos(&self, area: Rect) -> Option<(u16, u16)> {
-        self.bottom_pane.cursor_pos(area)
+        let [_, bottom_pane_area] = self.layout_areas(area);
+        self.bottom_pane.cursor_pos(bottom_pane_area)
     }
 }
 
 impl ChatWidget<'_> {
     fn begin_stream(&mut self, kind: StreamKind) {
+        if let Some(current) = self.current_stream {
+            if current != kind {
+                self.finalize_stream(current);
+            }
+        }
+
         if self.current_stream != Some(kind) {
             self.current_stream = Some(kind);
             self.stream_header_emitted = false;
@@ -526,6 +653,7 @@ impl ChatWidget<'_> {
             // Ensure the waiting status is visible (composer replaced).
             self.bottom_pane
                 .update_status_text("waiting for model".to_string());
+            self.emit_stream_header(kind);
         }
     }
 
@@ -609,10 +737,11 @@ impl ChatWidget<'_> {
 
 impl WidgetRef for &ChatWidget<'_> {
     fn render_ref(&self, area: Rect, buf: &mut Buffer) {
-        // In the hybrid inline viewport mode we only draw the interactive
-        // bottom pane; history entries are injected directly into scrollback
-        // via `Terminal::insert_before`.
-        (&self.bottom_pane).render(area, buf);
+        let [active_cell_area, bottom_pane_area] = self.layout_areas(area);
+        (&self.bottom_pane).render(bottom_pane_area, buf);
+        if let Some(cell) = &self.active_history_cell {
+            cell.render_ref(active_cell_area, buf);
+        }
     }
 }
 

codex-rs/tui/src/cli.rs

@@ -17,6 +17,12 @@ pub struct Cli {
     #[arg(long, short = 'm')]
     pub model: Option<String>,
 
+    /// Convenience flag to select the local open source model provider.
+    /// Equivalent to -c model_provider=oss; verifies a local Ollama server is
+    /// running.
+    #[arg(long = "oss", default_value_t = false)]
+    pub oss: bool,
+
     /// Configuration profile from config.toml to specify default options.
     #[arg(long = "profile", short = 'p')]
     pub config_profile: Option<String>,

codex-rs/tui/src/colors.rs

@@ -0,0 +1,4 @@
+use ratatui::style::Color;
+
+pub(crate) const LIGHT_BLUE: Color = Color::Rgb(134, 238, 255);
+pub(crate) const SUCCESS_GREEN: Color = Color::Rgb(169, 230, 158);

codex-rs/tui/src/history_cell.rs

@@ -1,19 +1,22 @@
+use crate::exec_command::relativize_to_home;
 use crate::exec_command::strip_bash_lc_and_escape;
+use crate::slash_command::SlashCommand;
 use crate::text_block::TextBlock;
 use crate::text_formatting::format_and_truncate_tool_result;
 use base64::Engine;
 use codex_ansi_escape::ansi_escape_line;
+use codex_common::create_config_summary_entries;
 use codex_common::elapsed::format_duration;
-use codex_common::summarize_sandbox_policy;
-use codex_core::WireApi;
 use codex_core::config::Config;
-use codex_core::model_supports_reasoning_summaries;
 use codex_core::plan_tool::PlanItemArg;
 use codex_core::plan_tool::StepStatus;
 use codex_core::plan_tool::UpdatePlanArgs;
+use codex_core::protocol::DiffHunk;
+use codex_core::protocol::DiffLineKind;
 use codex_core::protocol::FileChange;
 use codex_core::protocol::McpInvocation;
 use codex_core::protocol::SessionConfiguredEvent;
+use codex_core::protocol::TokenUsage;
 use image::DynamicImage;
 use image::ImageReader;
 use mcp_types::EmbeddedResourceResource;
@@ -24,6 +27,9 @@ use ratatui::style::Modifier;
 use ratatui::style::Style;
 use ratatui::text::Line as RtLine;
 use ratatui::text::Span as RtSpan;
+use ratatui::widgets::Paragraph;
+use ratatui::widgets::WidgetRef;
+use ratatui::widgets::Wrap;
 use std::collections::HashMap;
 use std::io::Cursor;
 use std::path::PathBuf;
@@ -96,6 +102,9 @@ pub(crate) enum HistoryCell {
     /// Output from the `/diff` command.
     GitDiffOutput { view: TextBlock },
 
+    /// Output from the `/status` command.
+    StatusOutput { view: TextBlock },
+
     /// Error event from the backend.
     ErrorEvent { view: TextBlock },
 
@@ -110,6 +119,9 @@ pub(crate) enum HistoryCell {
     /// A human‑friendly rendering of the model's current plan and step
     /// statuses provided via the `update_plan` tool.
     PlanUpdate { view: TextBlock },
+
+    /// Result of applying a patch (success or failure) with optional output.
+    PatchApplyResult { view: TextBlock },
 }
 
 const TOOL_CALL_MAX_LINES: usize = 5;
@@ -124,12 +136,14 @@ impl HistoryCell {
             | HistoryCell::UserPrompt { view }
             | HistoryCell::BackgroundEvent { view }
             | HistoryCell::GitDiffOutput { view }
+            | HistoryCell::StatusOutput { view }
             | HistoryCell::ErrorEvent { view }
             | HistoryCell::SessionInfo { view }
             | HistoryCell::CompletedExecCommand { view }
             | HistoryCell::CompletedMcpToolCall { view }
             | HistoryCell::PendingPatch { view }
             | HistoryCell::PlanUpdate { view }
+            | HistoryCell::PatchApplyResult { view }
             | HistoryCell::ActiveExecCommand { view, .. }
             | HistoryCell::ActiveMcpToolCall { view, .. } => {
                 view.lines.iter().map(line_to_static).collect()
@@ -140,6 +154,15 @@ impl HistoryCell {
             ],
         }
     }
+
+    pub(crate) fn desired_height(&self, width: u16) -> u16 {
+        Paragraph::new(Text::from(self.plain_lines()))
+            .wrap(Wrap { trim: false })
+            .line_count(width)
+            .try_into()
+            .unwrap_or(0)
+    }
+
     pub(crate) fn new_session_info(
         config: &Config,
         event: SessionConfiguredEvent,
@@ -147,51 +170,35 @@ impl HistoryCell {
     ) -> Self {
         let SessionConfiguredEvent {
             model,
-            session_id,
+            session_id: _,
             history_log_id: _,
             history_entry_count: _,
         } = event;
         if is_first_event {
-            const VERSION: &str = env!("CARGO_PKG_VERSION");
+            let cwd_str = match relativize_to_home(&config.cwd) {
+                Some(rel) if !rel.as_os_str().is_empty() => format!("~/{}", rel.display()),
+                Some(_) => "~".to_string(),
+                None => config.cwd.display().to_string(),
+            };
 
-            let mut lines: Vec<Line<'static>> = vec![
+            let lines: Vec<Line<'static>> = vec![
                 Line::from(vec![
-                    "OpenAI ".into(),
-                    "Codex".bold(),
-                    format!(" v{VERSION}").into(),
-                    " (research preview)".dim(),
-                ]),
-                Line::from(""),
-                Line::from(vec![
-                    "codex session".magenta().bold(),
-                    " ".into(),
-                    session_id.to_string().dim(),
+                    Span::raw(">_ ").dim(),
+                    Span::styled(
+                        "You are using OpenAI Codex in",
+                        Style::default().add_modifier(Modifier::BOLD),
+                    ),
+                    Span::raw(format!(" {cwd_str}")).dim(),
                 ]),
+                Line::from("".dim()),
+                Line::from(" Try one of the following commands to get started:".dim()),
+                Line::from("".dim()),
+                Line::from(format!(" 1. /init - {}", SlashCommand::Init.description()).dim()),
+                Line::from(format!(" 2. /status - {}", SlashCommand::Status.description()).dim()),
+                Line::from(format!(" 3. /compact - {}", SlashCommand::Compact.description()).dim()),
+                Line::from(format!(" 4. /new - {}", SlashCommand::New.description()).dim()),
+                Line::from("".dim()),
             ];
-
-            let mut entries = vec![
-                ("workdir", config.cwd.display().to_string()),
-                ("model", config.model.clone()),
-                ("provider", config.model_provider_id.clone()),
-                ("approval", config.approval_policy.to_string()),
-                ("sandbox", summarize_sandbox_policy(&config.sandbox_policy)),
-            ];
-            if config.model_provider.wire_api == WireApi::Responses
-                && model_supports_reasoning_summaries(config)
-            {
-                entries.push((
-                    "reasoning effort",
-                    config.model_reasoning_effort.to_string(),
-                ));
-                entries.push((
-                    "reasoning summaries",
-                    config.model_reasoning_summary.to_string(),
-                ));
-            }
-            for (key, value) in entries {
-                lines.push(Line::from(vec![format!("{key}: ").bold(), value.into()]));
-            }
-            lines.push(Line::from(""));
             HistoryCell::WelcomeMessage {
                 view: TextBlock::new(lines),
             }
@@ -445,6 +452,49 @@ impl HistoryCell {
         }
     }
 
+    pub(crate) fn new_status_output(config: &Config, usage: &TokenUsage) -> Self {
+        let mut lines: Vec<Line<'static>> = Vec::new();
+        lines.push(Line::from("/status".magenta()));
+
+        // Config
+        for (key, value) in create_config_summary_entries(config) {
+            lines.push(Line::from(vec![format!("{key}: ").bold(), value.into()]));
+        }
+
+        // Token usage
+        lines.push(Line::from(""));
+        lines.push(Line::from("token usage".bold()));
+        lines.push(Line::from(vec![
+            "  input: ".bold(),
+            usage.input_tokens.to_string().into(),
+        ]));
+        lines.push(Line::from(vec![
+            "  cached input: ".bold(),
+            usage.cached_input_tokens.unwrap_or(0).to_string().into(),
+        ]));
+        lines.push(Line::from(vec![
+            "  output: ".bold(),
+            usage.output_tokens.to_string().into(),
+        ]));
+        lines.push(Line::from(vec![
+            "  reasoning output: ".bold(),
+            usage
+                .reasoning_output_tokens
+                .unwrap_or(0)
+                .to_string()
+                .into(),
+        ]));
+        lines.push(Line::from(vec![
+            "  total: ".bold(),
+            usage.total_tokens.to_string().into(),
+        ]));
+
+        lines.push(Line::from(""));
+        HistoryCell::StatusOutput {
+            view: TextBlock::new(lines),
+        }
+    }
+
     pub(crate) fn new_error_event(message: String) -> Self {
         let lines: Vec<Line<'static>> = vec![
             vec!["ERROR: ".red().bold(), message.into()].into(),
@@ -551,7 +601,10 @@ impl HistoryCell {
             PatchEventType::ApplyBegin {
                 auto_approved: false,
             } => {
-                let lines = vec![Line::from("patch applied".magenta().bold())];
+                let lines: Vec<Line<'static>> = vec![
+                    Line::from("applying patch".magenta().bold()),
+                    Line::from(""),
+                ];
                 return Self::PendingPatch {
                     view: TextBlock::new(lines),
                 };
@@ -567,30 +620,7 @@ impl HistoryCell {
         lines.push(Line::from(title.magenta().bold()));
 
         for line in summary_lines {
-            if line.starts_with('+') {
-                lines.push(line.green().into());
-            } else if line.starts_with('-') {
-                lines.push(line.red().into());
-            } else if let Some(space_idx) = line.find(' ') {
-                let kind_owned = line[..space_idx].to_string();
-                let rest_owned = line[space_idx + 1..].to_string();
-
-                let style_for = |fg: Color| Style::default().fg(fg).add_modifier(Modifier::BOLD);
-
-                let styled_kind = match kind_owned.as_str() {
-                    "A" => RtSpan::styled(kind_owned.clone(), style_for(Color::Green)),
-                    "D" => RtSpan::styled(kind_owned.clone(), style_for(Color::Red)),
-                    "M" => RtSpan::styled(kind_owned.clone(), style_for(Color::Yellow)),
-                    "R" | "C" => RtSpan::styled(kind_owned.clone(), style_for(Color::Cyan)),
-                    _ => RtSpan::raw(kind_owned.clone()),
-                };
-
-                let styled_line =
-                    RtLine::from(vec![styled_kind, RtSpan::raw(" "), RtSpan::raw(rest_owned)]);
-                lines.push(styled_line);
-            } else {
-                lines.push(Line::from(line));
-            }
+            lines.push(line);
         }
 
         lines.push(Line::from(""));
@@ -599,33 +629,131 @@ impl HistoryCell {
             view: TextBlock::new(lines),
         }
     }
+
+    pub(crate) fn new_patch_apply_end(stdout: String, stderr: String, success: bool) -> Self {
+        let mut lines: Vec<Line<'static>> = Vec::new();
+
+        let status = if success {
+            RtSpan::styled("patch applied", Style::default().fg(Color::Green))
+        } else {
+            RtSpan::styled(
+                "patch failed",
+                Style::default().fg(Color::Red).add_modifier(Modifier::BOLD),
+            )
+        };
+        lines.push(RtLine::from(vec![
+            "patch".magenta().bold(),
+            " ".into(),
+            status,
+        ]));
+
+        let src = if success {
+            if stdout.trim().is_empty() {
+                &stderr
+            } else {
+                &stdout
+            }
+        } else if stderr.trim().is_empty() {
+            &stdout
+        } else {
+            &stderr
+        };
+
+        if !src.trim().is_empty() {
+            lines.push(Line::from(""));
+            let mut iter = src.lines();
+            for raw in iter.by_ref().take(TOOL_CALL_MAX_LINES) {
+                lines.push(ansi_escape_line(raw).dim());
+            }
+            let remaining = iter.count();
+            if remaining > 0 {
+                lines.push(Line::from(format!("... {remaining} additional lines")).dim());
+            }
+        }
+
+        lines.push(Line::from(""));
+
+        HistoryCell::PatchApplyResult {
+            view: TextBlock::new(lines),
+        }
+    }
 }
 
-fn create_diff_summary(changes: HashMap<PathBuf, FileChange>) -> Vec<String> {
+impl WidgetRef for &HistoryCell {
+    fn render_ref(&self, area: Rect, buf: &mut Buffer) {
+        Paragraph::new(Text::from(self.plain_lines()))
+            .wrap(Wrap { trim: false })
+            .render(area, buf);
+    }
+}
+
+fn create_diff_summary(changes: HashMap<PathBuf, FileChange>) -> Vec<Line<'static>> {
     // Build a concise, human‑readable summary list similar to the
     // `git status` short format so the user can reason about the
     // patch without scrolling.
-    let mut summaries: Vec<String> = Vec::new();
+    let mut summaries: Vec<Line<'static>> = Vec::new();
     for (path, change) in &changes {
         use codex_core::protocol::FileChange::*;
         match change {
             Add { content } => {
                 let added = content.lines().count();
-                summaries.push(format!("A {} (+{added})", path.display()));
+                summaries.push(Line::from(vec![
+                    RtSpan::styled(
+                        "A",
+                        Style::default()
+                            .fg(Color::Green)
+                            .add_modifier(Modifier::BOLD),
+                    ),
+                    RtSpan::raw(" "),
+                    RtSpan::raw(format!("{} (+{added})", path.display())),
+                ]));
             }
             Delete => {
-                summaries.push(format!("D {}", path.display()));
+                summaries.push(Line::from(vec![
+                    RtSpan::styled(
+                        "D",
+                        Style::default().fg(Color::Red).add_modifier(Modifier::BOLD),
+                    ),
+                    RtSpan::raw(" "),
+                    RtSpan::raw(format!("{}", path.display())),
+                ]));
             }
             Update {
                 unified_diff,
                 move_path,
+                hunks,
             } => {
                 if let Some(new_path) = move_path {
-                    summaries.push(format!("R {} → {}", path.display(), new_path.display(),));
+                    summaries.push(Line::from(vec![
+                        RtSpan::styled(
+                            "R",
+                            Style::default()
+                                .fg(Color::Cyan)
+                                .add_modifier(Modifier::BOLD),
+                        ),
+                        RtSpan::raw(" "),
+                        RtSpan::raw(format!("{} → {}", path.display(), new_path.display())),
+                    ]));
                 } else {
-                    summaries.push(format!("M {}", path.display(),));
+                    summaries.push(Line::from(vec![
+                        RtSpan::styled(
+                            "M",
+                            Style::default()
+                                .fg(Color::Yellow)
+                                .add_modifier(Modifier::BOLD),
+                        ),
+                        RtSpan::raw(" "),
+                        RtSpan::raw(format!("{}", path.display())),
+                    ]));
+                }
+
+                // Prefer structured hunks if provided; otherwise, fall back
+                // to parsing the unified diff string.
+                if let Some(hunks) = hunks {
+                    summaries.extend(format_inline_hunks(hunks));
+                } else {
+                    summaries.extend(format_inline_diff(unified_diff));
                 }
-                summaries.extend(unified_diff.lines().map(|s| s.to_string()));
             }
         }
     }
@@ -633,6 +761,246 @@ fn create_diff_summary(changes: HashMap<PathBuf, FileChange>) -> Vec<String> {
     summaries
 }
 
+/// Convert a unified diff (only hunk bodies expected) into inline diff lines
+/// with old/new line numbers and colored +/- content.
+fn format_inline_diff(unified_diff: &str) -> Vec<Line<'static>> {
+    // First pass: compute max widths for old/new line numbers from hunk headers
+    let mut max_old: usize = 0;
+    let mut max_new: usize = 0;
+    for l in unified_diff.lines() {
+        if let Some(caps) = parse_hunk_header(l) {
+            let (o_start, o_count, n_start, n_count) = caps;
+            let o_end = o_start + o_count.saturating_sub(1);
+            let n_end = n_start + n_count.saturating_sub(1);
+            max_old = max_old.max(o_end as usize);
+            max_new = max_new.max(n_end as usize);
+        }
+    }
+    let old_w = std::cmp::max(2, num_width(max_old as u64));
+    let new_w = std::cmp::max(2, num_width(max_new as u64));
+
+    let mut old_ln: i64 = 0;
+    let mut new_ln: i64 = 0;
+    let mut out: Vec<Line<'static>> = Vec::new();
+
+    for raw in unified_diff.lines() {
+        if let Some((o_start, _o_count, n_start, _n_count)) = parse_hunk_header(raw) {
+            old_ln = o_start as i64;
+            new_ln = n_start as i64;
+            out.push(Line::from(RtSpan::styled(
+                raw.to_string(),
+                Style::default().fg(Color::Cyan),
+            )));
+            continue;
+        }
+
+        // Prepare number columns and content coloring based on first char
+        let (left_num, right_num, content_style, advance_old, advance_new): (
+            String,
+            String,
+            Style,
+            i64,
+            i64,
+        ) = if raw.starts_with('+') {
+            (
+                " ".repeat(old_w),
+                format!("{:>width$}", new_ln, width = new_w),
+                Style::default().fg(Color::Green),
+                0,
+                1,
+            )
+        } else if raw.starts_with('-') {
+            (
+                format!("{:>width$}", old_ln, width = old_w),
+                " ".repeat(new_w),
+                Style::default().fg(Color::Red),
+                1,
+                0,
+            )
+        } else {
+            (
+                format!("{:>width$}", old_ln, width = old_w),
+                format!("{:>width$}", new_ln, width = new_w),
+                Style::default(),
+                1,
+                1,
+            )
+        };
+
+        let num_style = Style::default().fg(Color::DarkGray);
+        let sep = RtSpan::styled(" | ", num_style);
+        let left = RtSpan::styled(left_num, num_style);
+        let right = RtSpan::styled(right_num, num_style);
+        let content = RtSpan::styled(raw.to_string(), content_style);
+        out.push(RtLine::from(vec![left, sep.clone(), right, sep, content]));
+
+        old_ln += advance_old;
+        new_ln += advance_new;
+    }
+
+    out
+}
+
+/// Render structured hunks into inline diff with numbered columns, similar to
+/// `format_inline_diff` but without parsing unified diff text.
+fn format_inline_hunks(hunks: &[DiffHunk]) -> Vec<Line<'static>> {
+    // Compute max widths for columns based on hunk ranges.
+    let mut max_old_end: u32 = 0;
+    let mut max_new_end: u32 = 0;
+    for h in hunks {
+        let o_end = h.old_start.saturating_add(h.old_count.saturating_sub(1));
+        let n_end = h.new_start.saturating_add(h.new_count.saturating_sub(1));
+        max_old_end = max_old_end.max(o_end);
+        max_new_end = max_new_end.max(n_end);
+    }
+    let old_w = std::cmp::max(2, num_width(max_old_end as u64));
+    let new_w = std::cmp::max(2, num_width(max_new_end as u64));
+
+    let mut out: Vec<Line<'static>> = Vec::new();
+    let num_style = Style::default().fg(Color::DarkGray);
+
+    for h in hunks {
+        // Hunk header line
+        let header = format!(
+            "@@ -{},{} +{},{} @@",
+            h.old_start, h.old_count, h.new_start, h.new_count
+        );
+        out.push(Line::from(RtSpan::styled(header, Style::default().fg(Color::Cyan))));
+
+        let mut old_ln = h.old_start as i64;
+        let mut new_ln = h.new_start as i64;
+
+        for l in &h.lines {
+            let (left_num, right_num, content_style, advance_old, advance_new, prefix) =
+                match l.kind {
+                    DiffLineKind::Add => (
+                        " ".repeat(old_w),
+                        format!("{:>width$}", new_ln, width = new_w),
+                        Style::default().fg(Color::Green),
+                        0,
+                        1,
+                        '+',
+                    ),
+                    DiffLineKind::Delete => (
+                        format!("{:>width$}", old_ln, width = old_w),
+                        " ".repeat(new_w),
+                        Style::default().fg(Color::Red),
+                        1,
+                        0,
+                        '-',
+                    ),
+                    DiffLineKind::Context => (
+                        format!("{:>width$}", old_ln, width = old_w),
+                        format!("{:>width$}", new_ln, width = new_w),
+                        Style::default(),
+                        1,
+                        1,
+                        ' ',
+                    ),
+                };
+
+            let sep = RtSpan::styled(" | ", num_style);
+            let left = RtSpan::styled(left_num, num_style);
+            let right = RtSpan::styled(right_num, num_style);
+            let mut content = String::with_capacity(l.text.len() + 1);
+            content.push(prefix);
+            content.push_str(&l.text);
+            let content = RtSpan::styled(content, content_style);
+            out.push(RtLine::from(vec![left, sep.clone(), right, sep, content]));
+
+            old_ln += advance_old;
+            new_ln += advance_new;
+        }
+    }
+
+    out
+}
+
+fn parse_hunk_header(line: &str) -> Option<(u64, u64, u64, u64)> {
+    // @@ -oldStart,oldCount +newStart,newCount @@
+    // counts may be omitted which implies 1
+    if !line.starts_with("@@") {
+        return None;
+    }
+    let mut old_start: u64 = 0;
+    let mut old_count: u64 = 1;
+    let mut new_start: u64 = 0;
+    let mut new_count: u64 = 1;
+    // A simple, robust parse without regex
+    // Find "-" then space then "+" then space then "@@"
+    let bytes = line.as_bytes();
+    // Find first '-' after '@@'
+    let mut i = 2;
+    while i < bytes.len() && bytes[i].is_ascii_whitespace() {
+        i += 1;
+    }
+    if i >= bytes.len() || bytes[i] != b'-' {
+        return None;
+    }
+    i += 1;
+    // parse oldStart
+    let (n, consumed) = parse_uint(&bytes[i..]);
+    if consumed == 0 {
+        return None;
+    }
+    old_start = n;
+    i += consumed;
+    // optional ,oldCount
+    if i < bytes.len() && bytes[i] == b',' {
+        i += 1;
+        let (n2, c2) = parse_uint(&bytes[i..]);
+        if c2 == 0 {
+            return None;
+        }
+        old_count = n2;
+        i += c2;
+    }
+    // skip spaces
+    while i < bytes.len() && bytes[i].is_ascii_whitespace() {
+        i += 1;
+    }
+    if i >= bytes.len() || bytes[i] != b'+' {
+        return None;
+    }
+    i += 1;
+    let (n3, c3) = parse_uint(&bytes[i..]);
+    if c3 == 0 {
+        return None;
+    }
+    new_start = n3;
+    i += c3;
+    if i < bytes.len() && bytes[i] == b',' {
+        i += 1;
+        let (n4, c4) = parse_uint(&bytes[i..]);
+        if c4 == 0 {
+            return None;
+        }
+        new_count = n4;
+        i += c4;
+    }
+    Some((old_start, old_count, new_start, new_count))
+}
+
+fn parse_uint(s: &[u8]) -> (u64, usize) {
+    let mut i = 0usize;
+    let mut n: u64 = 0;
+    while i < s.len() {
+        let b = s[i];
+        if b.is_ascii_digit() {
+            n = n * 10 + (b - b'0') as u64;
+            i += 1;
+        } else {
+            break;
+        }
+    }
+    (n, i)
+}
+
+fn num_width(n: u64) -> usize {
+    // Simple and exact decimal width
+    n.to_string().len()
+}
+
 fn format_mcp_invocation<'a>(invocation: McpInvocation) -> Line<'a> {
     let args_str = invocation
         .arguments

codex-rs/tui/src/lib.rs

@@ -3,15 +3,16 @@
 // alternate‑screen mode starts; that file opts‑out locally via `allow`.
 #![deny(clippy::print_stdout, clippy::print_stderr)]
 use app::App;
+use codex_core::BUILT_IN_OSS_MODEL_PROVIDER_ID;
 use codex_core::config::Config;
 use codex_core::config::ConfigOverrides;
 use codex_core::config_types::SandboxMode;
 use codex_core::protocol::AskForApproval;
 use codex_core::util::is_inside_git_repo;
 use codex_login::load_auth;
+use codex_ollama::DEFAULT_OSS_MODEL;
 use log_layer::TuiLogLayer;
 use std::fs::OpenOptions;
-use std::io::Write;
 use std::path::PathBuf;
 use tracing::error;
 use tracing_appender::non_blocking;
@@ -25,6 +26,7 @@ mod bottom_pane;
 mod chatwidget;
 mod citation_regex;
 mod cli;
+mod colors;
 pub mod custom_terminal;
 mod exec_command;
 mod file_search;
@@ -35,6 +37,8 @@ pub mod insert_history;
 pub mod live_wrap;
 mod log_layer;
 mod markdown;
+pub mod onboarding;
+mod shimmer;
 mod slash_command;
 mod status_indicator_widget;
 mod text_block;
@@ -70,18 +74,37 @@ pub async fn run_main(
         )
     };
 
+    // When using `--oss`, let the bootstrapper pick the model (defaulting to
+    // gpt-oss:20b) and ensure it is present locally. Also, force the built‑in
+    // `oss` model provider.
+    let model = if let Some(model) = &cli.model {
+        Some(model.clone())
+    } else if cli.oss {
+        Some(DEFAULT_OSS_MODEL.to_owned())
+    } else {
+        None // No model specified, will use the default.
+    };
+
+    let model_provider_override = if cli.oss {
+        Some(BUILT_IN_OSS_MODEL_PROVIDER_ID.to_owned())
+    } else {
+        None
+    };
+
     let config = {
         // Load configuration and support CLI overrides.
         let overrides = ConfigOverrides {
-            model: cli.model.clone(),
+            model,
             approval_policy,
             sandbox_mode,
             cwd: cli.cwd.clone().map(|p| p.canonicalize().unwrap_or(p)),
-            model_provider: None,
+            model_provider: model_provider_override,
             config_profile: cli.config_profile.clone(),
             codex_linux_sandbox_exe,
             base_instructions: None,
             include_plan_tool: Some(true),
+            disable_response_storage: cli.oss.then_some(true),
+            show_raw_agent_reasoning: cli.oss.then_some(true),
         };
         // Parse `-c` overrides from the CLI.
         let cli_kv_overrides = match cli.config_overrides.parse_overrides() {
@@ -136,6 +159,12 @@ pub async fn run_main(
         .with_target(false)
         .with_filter(env_filter());
 
+    if cli.oss {
+        codex_ollama::ensure_oss_ready(&config)
+            .await
+            .map_err(|e| std::io::Error::other(format!("OSS setup failed: {e}")))?;
+    }
+
     // Channel that carries formatted log lines to the UI.
     let (log_tx, log_rx) = tokio::sync::mpsc::unbounded_channel::<String>();
     let tui_layer = TuiLogLayer::new(log_tx.clone(), 120).with_filter(env_filter());
@@ -177,24 +206,6 @@ pub async fn run_main(
         eprintln!("");
     }
 
-    let show_login_screen = should_show_login_screen(&config);
-    if show_login_screen {
-        std::io::stdout()
-            .write_all(b"No API key detected.\nLogin with your ChatGPT account? [Yn] ")?;
-        std::io::stdout().flush()?;
-        let mut input = String::new();
-        std::io::stdin().read_line(&mut input)?;
-        let trimmed = input.trim();
-        if !(trimmed.is_empty() || trimmed.eq_ignore_ascii_case("y")) {
-            std::process::exit(1);
-        }
-        // Spawn a task to run the login command.
-        // Block until the login command is finished.
-        codex_login::login_with_chatgpt(&config.codex_home, false).await?;
-
-        std::io::stdout().write_all(b"Login successful.\n")?;
-    }
-
     // Determine whether we need to display the "not a git repo" warning
     // modal. The flag is shown when the current working directory is *not*
     // inside a Git repository **and** the user did *not* pass the
@@ -260,7 +271,7 @@ fn restore() {
 
 #[allow(clippy::unwrap_used)]
 fn should_show_login_screen(config: &Config) -> bool {
-    if config.model_provider.requires_auth {
+    if config.model_provider.requires_openai_auth {
         // Reading the OpenAI API key is an async operation because it may need
         // to refresh the token. Block on it.
         let codex_home = config.codex_home.clone();

codex-rs/tui/src/main.rs

@@ -22,7 +22,9 @@ fn main() -> anyhow::Result<()> {
             .raw_overrides
             .splice(0..0, top_cli.config_overrides.raw_overrides);
         let usage = run_main(inner, codex_linux_sandbox_exe).await?;
-        println!("{}", codex_core::protocol::FinalOutput::from(usage));
+        if !usage.is_zero() {
+            println!("{}", codex_core::protocol::FinalOutput::from(usage));
+        }
         Ok(())
     })
 }

codex-rs/tui/src/onboarding/auth.rs

@@ -0,0 +1,316 @@
+use crossterm::event::KeyCode;
+use crossterm::event::KeyEvent;
+use ratatui::buffer::Buffer;
+use ratatui::layout::Rect;
+use ratatui::prelude::Widget;
+use ratatui::style::Color;
+use ratatui::style::Modifier;
+use ratatui::style::Style;
+use ratatui::text::Line;
+use ratatui::text::Span;
+use ratatui::widgets::Paragraph;
+use ratatui::widgets::WidgetRef;
+use ratatui::widgets::Wrap;
+
+use codex_login::AuthMode;
+
+use crate::app_event::AppEvent;
+use crate::app_event_sender::AppEventSender;
+use crate::colors::LIGHT_BLUE;
+use crate::colors::SUCCESS_GREEN;
+use crate::onboarding::onboarding_screen::KeyEventResult;
+use crate::onboarding::onboarding_screen::KeyboardHandler;
+use crate::shimmer::FrameTicker;
+use crate::shimmer::shimmer_spans;
+use std::path::PathBuf;
+// no additional imports
+
+#[derive(Debug)]
+pub(crate) enum SignInState {
+    PickMode,
+    ChatGptContinueInBrowser(#[allow(dead_code)] ContinueInBrowserState),
+    ChatGptSuccess,
+}
+
+#[derive(Debug)]
+/// Used to manage the lifecycle of SpawnedLogin and FrameTicker and ensure they get cleaned up.
+pub(crate) struct ContinueInBrowserState {
+    _login_child: Option<codex_login::SpawnedLogin>,
+    _frame_ticker: Option<FrameTicker>,
+}
+
+impl Drop for ContinueInBrowserState {
+    fn drop(&mut self) {
+        if let Some(child) = &self._login_child {
+            if let Ok(mut locked) = child.child.lock() {
+                // Best-effort terminate and reap the child to avoid zombies.
+                let _ = locked.kill();
+                let _ = locked.wait();
+            }
+        }
+    }
+}
+
+impl KeyboardHandler for AuthModeWidget {
+    fn handle_key_event(&mut self, key_event: KeyEvent) -> KeyEventResult {
+        match key_event.code {
+            KeyCode::Up | KeyCode::Char('k') => {
+                self.mode = AuthMode::ChatGPT;
+                KeyEventResult::None
+            }
+            KeyCode::Down | KeyCode::Char('j') => {
+                self.mode = AuthMode::ApiKey;
+                KeyEventResult::None
+            }
+            KeyCode::Char('1') => {
+                self.mode = AuthMode::ChatGPT;
+                self.start_chatgpt_login();
+                KeyEventResult::None
+            }
+            KeyCode::Char('2') => {
+                self.mode = AuthMode::ApiKey;
+                self.verify_api_key()
+            }
+            KeyCode::Enter => match self.mode {
+                AuthMode::ChatGPT => match &self.sign_in_state {
+                    SignInState::PickMode => self.start_chatgpt_login(),
+                    SignInState::ChatGptContinueInBrowser(_) => KeyEventResult::None,
+                    SignInState::ChatGptSuccess => KeyEventResult::Continue,
+                },
+                AuthMode::ApiKey => self.verify_api_key(),
+            },
+            KeyCode::Esc => {
+                if matches!(self.sign_in_state, SignInState::ChatGptContinueInBrowser(_)) {
+                    self.sign_in_state = SignInState::PickMode;
+                    self.event_tx.send(AppEvent::RequestRedraw);
+                    KeyEventResult::None
+                } else {
+                    KeyEventResult::Quit
+                }
+            }
+            KeyCode::Char('q') => KeyEventResult::Quit,
+            _ => KeyEventResult::None,
+        }
+    }
+}
+
+#[derive(Debug)]
+pub(crate) struct AuthModeWidget {
+    pub mode: AuthMode,
+    pub error: Option<String>,
+    pub sign_in_state: SignInState,
+    pub event_tx: AppEventSender,
+    pub codex_home: PathBuf,
+}
+
+impl AuthModeWidget {
+    fn render_pick_mode(&self, area: Rect, buf: &mut Buffer) {
+        let mut lines: Vec<Line> = vec![
+            Line::from(vec![
+                Span::raw("> "),
+                Span::styled(
+                    "Sign in with your ChatGPT account?",
+                    Style::default().add_modifier(Modifier::BOLD),
+                ),
+            ]),
+            Line::from(""),
+        ];
+
+        let create_mode_item = |idx: usize,
+                                selected_mode: AuthMode,
+                                text: &str,
+                                description: &str|
+         -> Vec<Line<'static>> {
+            let is_selected = self.mode == selected_mode;
+            let caret = if is_selected { ">" } else { " " };
+
+            let line1 = if is_selected {
+                Line::from(vec![
+                    Span::styled(
+                        format!("{} {}. ", caret, idx + 1),
+                        Style::default().fg(LIGHT_BLUE).add_modifier(Modifier::DIM),
+                    ),
+                    Span::styled(text.to_owned(), Style::default().fg(LIGHT_BLUE)),
+                ])
+            } else {
+                Line::from(format!("  {}. {text}", idx + 1))
+            };
+
+            let line2 = if is_selected {
+                Line::from(format!("     {description}"))
+                    .style(Style::default().fg(LIGHT_BLUE).add_modifier(Modifier::DIM))
+            } else {
+                Line::from(format!("     {description}"))
+                    .style(Style::default().add_modifier(Modifier::DIM))
+            };
+
+            vec![line1, line2]
+        };
+
+        lines.extend(create_mode_item(
+            0,
+            AuthMode::ChatGPT,
+            "Sign in with ChatGPT or create a new account",
+            "Leverages your plan, starting at $20 a month for Plus",
+        ));
+        lines.extend(create_mode_item(
+            1,
+            AuthMode::ApiKey,
+            "Provide your own API key",
+            "Pay only for what you use",
+        ));
+        lines.push(Line::from(""));
+        lines.push(
+            Line::from("Press Enter to continue")
+                .style(Style::default().add_modifier(Modifier::DIM)),
+        );
+        if let Some(err) = &self.error {
+            lines.push(Line::from(""));
+            lines.push(Line::from(Span::styled(
+                err.as_str(),
+                Style::default().fg(Color::Red),
+            )));
+        }
+
+        Paragraph::new(lines)
+            .wrap(Wrap { trim: false })
+            .render(area, buf);
+    }
+
+    fn render_continue_in_browser(&self, area: Rect, buf: &mut Buffer) {
+        let idx = self.current_frame();
+        let mut spans = vec![Span::from("> ")];
+        spans.extend(shimmer_spans("Finish signing in via your browser", idx));
+        let lines = vec![
+            Line::from(spans),
+            Line::from(""),
+            Line::from("  Press Escape to cancel")
+                .style(Style::default().add_modifier(Modifier::DIM)),
+        ];
+        Paragraph::new(lines)
+            .wrap(Wrap { trim: false })
+            .render(area, buf);
+    }
+
+    fn render_chatgpt_success(&self, area: Rect, buf: &mut Buffer) {
+        let lines = vec![
+            Line::from("✓ Signed in with your ChatGPT account")
+                .style(Style::default().fg(SUCCESS_GREEN)),
+            Line::from(""),
+            Line::from("> Before you start:"),
+            Line::from(""),
+            Line::from("  Codex can make mistakes"),
+            Line::from("  Check important info")
+                .style(Style::default().add_modifier(Modifier::DIM)),
+            Line::from(""),
+            Line::from("  Due to prompt injection risks, only use it with code you trust"),
+            Line::from("  For more details see https://github.com/openai/codex")
+                .style(Style::default().add_modifier(Modifier::DIM)),
+            Line::from(""),
+            Line::from("  Powered by your ChatGPT account"),
+            Line::from("  Uses your plan's rate limits and training data preferences")
+                .style(Style::default().add_modifier(Modifier::DIM)),
+            Line::from(""),
+            Line::from("  Press Enter to continue").style(Style::default().fg(LIGHT_BLUE)),
+        ];
+
+        Paragraph::new(lines)
+            .wrap(Wrap { trim: false })
+            .render(area, buf);
+    }
+
+    fn start_chatgpt_login(&mut self) -> KeyEventResult {
+        self.error = None;
+        match codex_login::spawn_login_with_chatgpt(&self.codex_home) {
+            Ok(child) => {
+                self.spawn_completion_poller(child.clone());
+                self.sign_in_state =
+                    SignInState::ChatGptContinueInBrowser(ContinueInBrowserState {
+                        _login_child: Some(child),
+                        _frame_ticker: Some(FrameTicker::new(self.event_tx.clone())),
+                    });
+                self.event_tx.send(AppEvent::RequestRedraw);
+                KeyEventResult::None
+            }
+            Err(e) => {
+                self.sign_in_state = SignInState::PickMode;
+                self.error = Some(e.to_string());
+                self.event_tx.send(AppEvent::RequestRedraw);
+                KeyEventResult::None
+            }
+        }
+    }
+
+    /// TODO: Read/write from the correct hierarchy config overrides + auth json + OPENAI_API_KEY.
+    fn verify_api_key(&mut self) -> KeyEventResult {
+        if std::env::var("OPENAI_API_KEY").is_err() {
+            self.error =
+                Some("Set OPENAI_API_KEY in your environment. Learn more: https://platform.openai.com/docs/libraries".to_string());
+            self.event_tx.send(AppEvent::RequestRedraw);
+            KeyEventResult::None
+        } else {
+            KeyEventResult::Continue
+        }
+    }
+
+    fn spawn_completion_poller(&self, child: codex_login::SpawnedLogin) {
+        let child_arc = child.child.clone();
+        let stderr_buf = child.stderr.clone();
+        let event_tx = self.event_tx.clone();
+        std::thread::spawn(move || {
+            loop {
+                let done = {
+                    if let Ok(mut locked) = child_arc.lock() {
+                        match locked.try_wait() {
+                            Ok(Some(status)) => Some(status.success()),
+                            Ok(None) => None,
+                            Err(_) => Some(false),
+                        }
+                    } else {
+                        Some(false)
+                    }
+                };
+                if let Some(success) = done {
+                    if success {
+                        event_tx.send(AppEvent::OnboardingAuthComplete(Ok(())));
+                    } else {
+                        let err = stderr_buf
+                            .lock()
+                            .ok()
+                            .and_then(|b| String::from_utf8(b.clone()).ok())
+                            .unwrap_or_else(|| "login_with_chatgpt subprocess failed".to_string());
+                        event_tx.send(AppEvent::OnboardingAuthComplete(Err(err)));
+                    }
+                    break;
+                }
+                std::thread::sleep(std::time::Duration::from_millis(250));
+            }
+        });
+    }
+
+    fn current_frame(&self) -> usize {
+        // Derive frame index from wall-clock time to avoid storing animation state.
+        // 100ms per frame to match the previous ticker cadence.
+        let now_ms = std::time::SystemTime::now()
+            .duration_since(std::time::UNIX_EPOCH)
+            .map(|d| d.as_millis())
+            .unwrap_or(0);
+        (now_ms / 100) as usize
+    }
+}
+
+impl WidgetRef for AuthModeWidget {
+    fn render_ref(&self, area: Rect, buf: &mut Buffer) {
+        match self.sign_in_state {
+            SignInState::PickMode => {
+                self.render_pick_mode(area, buf);
+            }
+            SignInState::ChatGptContinueInBrowser(_) => {
+                self.render_continue_in_browser(area, buf);
+            }
+            SignInState::ChatGptSuccess => {
+                self.render_chatgpt_success(area, buf);
+            }
+        }
+    }
+}

codex-rs/tui/src/onboarding/mod.rs

@@ -0,0 +1,3 @@
+mod auth;
+pub mod onboarding_screen;
+mod welcome;

codex-rs/tui/src/onboarding/onboarding_screen.rs

@@ -0,0 +1,157 @@
+use crossterm::event::KeyEvent;
+use ratatui::buffer::Buffer;
+use ratatui::layout::Rect;
+use ratatui::widgets::WidgetRef;
+
+use codex_login::AuthMode;
+
+use crate::app_event::AppEvent;
+use crate::app_event_sender::AppEventSender;
+use crate::onboarding::auth::AuthModeWidget;
+use crate::onboarding::auth::SignInState;
+use crate::onboarding::welcome::WelcomeWidget;
+use std::path::PathBuf;
+
+enum Step {
+    Welcome(WelcomeWidget),
+    Auth(AuthModeWidget),
+}
+
+pub(crate) trait KeyboardHandler {
+    fn handle_key_event(&mut self, key_event: KeyEvent) -> KeyEventResult;
+}
+
+pub(crate) enum KeyEventResult {
+    Continue,
+    Quit,
+    None,
+}
+
+pub(crate) struct OnboardingScreen {
+    event_tx: AppEventSender,
+    steps: Vec<Step>,
+}
+
+impl OnboardingScreen {
+    pub(crate) fn new(event_tx: AppEventSender, codex_home: PathBuf) -> Self {
+        let steps: Vec<Step> = vec![
+            Step::Welcome(WelcomeWidget {}),
+            Step::Auth(AuthModeWidget {
+                event_tx: event_tx.clone(),
+                mode: AuthMode::ChatGPT,
+                error: None,
+                sign_in_state: SignInState::PickMode,
+                codex_home,
+            }),
+        ];
+        Self { event_tx, steps }
+    }
+
+    pub(crate) fn on_auth_complete(&mut self, result: Result<(), String>) -> KeyEventResult {
+        if let Some(Step::Auth(state)) = self.steps.last_mut() {
+            match result {
+                Ok(()) => {
+                    state.sign_in_state = SignInState::ChatGptSuccess;
+                    self.event_tx.send(AppEvent::RequestRedraw);
+                    KeyEventResult::None
+                }
+                Err(e) => {
+                    state.sign_in_state = SignInState::PickMode;
+                    state.error = Some(e);
+                    self.event_tx.send(AppEvent::RequestRedraw);
+                    KeyEventResult::None
+                }
+            }
+        } else {
+            KeyEventResult::None
+        }
+    }
+}
+
+impl KeyboardHandler for OnboardingScreen {
+    fn handle_key_event(&mut self, key_event: KeyEvent) -> KeyEventResult {
+        if let Some(last_step) = self.steps.last_mut() {
+            self.event_tx.send(AppEvent::RequestRedraw);
+            last_step.handle_key_event(key_event)
+        } else {
+            KeyEventResult::None
+        }
+    }
+}
+
+impl WidgetRef for &OnboardingScreen {
+    fn render_ref(&self, area: Rect, buf: &mut Buffer) {
+        // Render steps top-to-bottom, measuring each step's height dynamically.
+        let mut y = area.y;
+        let bottom = area.y.saturating_add(area.height);
+        let width = area.width;
+
+        // Helper to scan a temporary buffer and return number of used rows.
+        fn used_rows(tmp: &Buffer, width: u16, height: u16) -> u16 {
+            if width == 0 || height == 0 {
+                return 0;
+            }
+            let mut last_non_empty: Option<u16> = None;
+            for yy in 0..height {
+                let mut any = false;
+                for xx in 0..width {
+                    let sym = tmp[(xx, yy)].symbol();
+                    if !sym.trim().is_empty() {
+                        any = true;
+                        break;
+                    }
+                }
+                if any {
+                    last_non_empty = Some(yy);
+                }
+            }
+            last_non_empty.map(|v| v + 2).unwrap_or(0)
+        }
+
+        let mut i = 0usize;
+        while i < self.steps.len() && y < bottom {
+            let step = &self.steps[i];
+            let max_h = bottom.saturating_sub(y);
+            if max_h == 0 || width == 0 {
+                break;
+            }
+            let scratch_area = Rect::new(0, 0, width, max_h);
+            let mut scratch = Buffer::empty(scratch_area);
+            step.render_ref(scratch_area, &mut scratch);
+            let h = used_rows(&scratch, width, max_h).min(max_h);
+            if h > 0 {
+                let target = Rect {
+                    x: area.x,
+                    y,
+                    width,
+                    height: h,
+                };
+                step.render_ref(target, buf);
+                y = y.saturating_add(h);
+            }
+            i += 1;
+        }
+    }
+}
+
+impl KeyboardHandler for Step {
+    fn handle_key_event(&mut self, key_event: KeyEvent) -> KeyEventResult {
+        match self {
+            Step::Welcome(_) => KeyEventResult::None,
+            Step::Auth(widget) => widget.handle_key_event(key_event),
+        }
+    }
+}
+
+impl WidgetRef for Step {
+    fn render_ref(&self, area: Rect, buf: &mut Buffer) {
+        match self {
+            Step::Welcome(widget) => {
+                widget.render_ref(area, buf);
+            }
+            Step::Auth(widget) => {
+                widget.render_ref(area, buf);
+            }
+        }
+    }
+}

codex-rs/tui/src/onboarding/welcome.rs

@@ -0,0 +1,23 @@
+use ratatui::buffer::Buffer;
+use ratatui::layout::Rect;
+use ratatui::prelude::Widget;
+use ratatui::style::Modifier;
+use ratatui::style::Style;
+use ratatui::text::Line;
+use ratatui::text::Span;
+use ratatui::widgets::WidgetRef;
+
+pub(crate) struct WelcomeWidget {}
+
+impl WidgetRef for &WelcomeWidget {
+    fn render_ref(&self, area: Rect, buf: &mut Buffer) {
+        let line = Line::from(vec![
+            Span::raw("> "),
+            Span::styled(
+                "Welcome to Codex, OpenAI's coding agent that runs in your terminal",
+                Style::default().add_modifier(Modifier::BOLD),
+            ),
+        ]);
+        line.render(area, buf);
+    }
+}

codex-rs/tui/src/shimmer.rs

@@ -0,0 +1,84 @@
+use std::sync::Arc;
+use std::sync::atomic::AtomicBool;
+use std::sync::atomic::Ordering;
+use std::time::Duration;
+
+use ratatui::style::Color;
+use ratatui::style::Modifier;
+use ratatui::style::Style;
+use ratatui::text::Span;
+
+use crate::app_event::AppEvent;
+use crate::app_event_sender::AppEventSender;
+
+#[derive(Debug)]
+pub(crate) struct FrameTicker {
+    running: Arc<AtomicBool>,
+}
+
+impl FrameTicker {
+    pub(crate) fn new(app_event_tx: AppEventSender) -> Self {
+        let running = Arc::new(AtomicBool::new(true));
+        let running_clone = running.clone();
+        let app_event_tx_clone = app_event_tx.clone();
+        std::thread::spawn(move || {
+            while running_clone.load(Ordering::Relaxed) {
+                std::thread::sleep(Duration::from_millis(100));
+                app_event_tx_clone.send(AppEvent::RequestRedraw);
+            }
+        });
+        Self { running }
+    }
+}
+
+impl Drop for FrameTicker {
+    fn drop(&mut self) {
+        self.running.store(false, Ordering::Relaxed);
+    }
+}
+
+pub(crate) fn shimmer_spans(text: &str, frame_idx: usize) -> Vec<Span<'static>> {
+    let chars: Vec<char> = text.chars().collect();
+    let padding = 10usize;
+    let period = chars.len() + padding * 2;
+    let pos = frame_idx % period;
+    let has_true_color = supports_color::on_cached(supports_color::Stream::Stdout)
+        .map(|level| level.has_16m)
+        .unwrap_or(false);
+    let band_half_width = 6.0;
+
+    let mut spans: Vec<Span<'static>> = Vec::with_capacity(chars.len());
+    for (i, ch) in chars.iter().enumerate() {
+        let i_pos = i as isize + padding as isize;
+        let pos = pos as isize;
+        let dist = (i_pos - pos).abs() as f32;
+
+        let t = if dist <= band_half_width {
+            let x = std::f32::consts::PI * (dist / band_half_width);
+            0.5 * (1.0 + x.cos())
+        } else {
+            0.0
+        };
+        let brightness = 0.4 + 0.6 * t;
+        let level = (brightness * 255.0).clamp(0.0, 255.0) as u8;
+        let style = if has_true_color {
+            Style::default()
+                .fg(Color::Rgb(level, level, level))
+                .add_modifier(Modifier::BOLD)
+        } else {
+            Style::default().fg(color_for_level(level))
+        };
+        spans.push(Span::styled(ch.to_string(), style));
+    }
+    spans
+}
+
+fn color_for_level(level: u8) -> Color {
+    if level < 128 {
+        Color::DarkGray
+    } else if level < 192 {
+        Color::Gray
+    } else {
+        Color::White
+    }
+}

codex-rs/tui/src/slash_command.rs

@@ -13,8 +13,10 @@ pub enum SlashCommand {
     // DO NOT ALPHA-SORT! Enum order is presentation order in the popup, so
     // more frequently used commands should be listed first.
     New,
+    Init,
     Compact,
     Diff,
+    Status,
     Quit,
     #[cfg(debug_assertions)]
     TestApproval,
@@ -24,12 +26,12 @@ impl SlashCommand {
     /// User-visible description shown in the popup.
     pub fn description(self) -> &'static str {
         match self {
-            SlashCommand::New => "Start a new chat.",
-            SlashCommand::Compact => "Compact the chat history.",
-            SlashCommand::Quit => "Exit the application.",
-            SlashCommand::Diff => {
-                "Show git diff of the working directory (including untracked files)"
-            }
+            SlashCommand::New => "Start a new chat",
+            SlashCommand::Init => "Create an AGENTS.md file with instructions for Codex",
+            SlashCommand::Compact => "Compact the chat history",
+            SlashCommand::Quit => "Exit the application",
+            SlashCommand::Diff => "Show git diff (including untracked files)",
+            SlashCommand::Status => "Show current session configuration and token usage",
             #[cfg(debug_assertions)]
             SlashCommand::TestApproval => "Test approval request",
         }

codex-rs/tui/src/tui.rs

@@ -3,11 +3,14 @@ use std::io::Stdout;
 use std::io::stdout;
 
 use codex_core::config::Config;
+use crossterm::cursor::MoveTo;
 use crossterm::event::DisableBracketedPaste;
 use crossterm::event::EnableBracketedPaste;
 use crossterm::event::KeyboardEnhancementFlags;
 use crossterm::event::PopKeyboardEnhancementFlags;
 use crossterm::event::PushKeyboardEnhancementFlags;
+use crossterm::terminal::Clear;
+use crossterm::terminal::ClearType;
 use ratatui::backend::CrosstermBackend;
 use ratatui::crossterm::execute;
 use ratatui::crossterm::terminal::disable_raw_mode;
@@ -36,6 +39,9 @@ pub fn init(_config: &Config) -> Result<Tui> {
     )?;
     set_panic_hook();
 
+    // Clear screen and move cursor to top-left before drawing UI
+    execute!(stdout(), Clear(ClearType::All), MoveTo(0, 0))?;
+
     let backend = CrosstermBackend::new(stdout());
     let tui = Terminal::with_options(backend)?;
     Ok(tui)

codex-rs/tui/src/user_approval_widget.rs

@@ -47,6 +47,8 @@ pub(crate) enum ApprovalRequest {
 }
 
 /// Options displayed in the *select* mode.
+///
+/// The `key` is matched case-insensitively.
 struct SelectOption {
     label: Line<'static>,
     description: &'static str,
@@ -187,6 +189,16 @@ impl UserApprovalWidget<'_> {
         }
     }
 
+    /// Normalize a key for comparison.
+    /// - For `KeyCode::Char`, converts to lowercase for case-insensitive matching.
+    /// - Other key codes are returned unchanged.
+    fn normalize_keycode(code: KeyCode) -> KeyCode {
+        match code {
+            KeyCode::Char(c) => KeyCode::Char(c.to_ascii_lowercase()),
+            other => other,
+        }
+    }
+
     /// Handle Ctrl-C pressed by the user while the modal is visible.
     /// Behaves like pressing Escape: abort the request and close the modal.
     pub(crate) fn on_ctrl_c(&mut self) {
@@ -210,7 +222,12 @@ impl UserApprovalWidget<'_> {
                 self.send_decision(ReviewDecision::Abort);
             }
             other => {
-                if let Some(opt) = self.select_options.iter().find(|opt| opt.key == other) {
+                let normalized = Self::normalize_keycode(other);
+                if let Some(opt) = self
+                    .select_options
+                    .iter()
+                    .find(|opt| Self::normalize_keycode(opt.key) == normalized)
+                {
                     self.send_decision(opt.decision);
                 }
             }
@@ -330,3 +347,59 @@ impl WidgetRef for &UserApprovalWidget<'_> {
             );
     }
 }
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+    use crossterm::event::KeyCode;
+    use crossterm::event::KeyEvent;
+    use crossterm::event::KeyModifiers;
+    use std::path::PathBuf;
+    use std::sync::mpsc::channel;
+
+    #[test]
+    fn lowercase_shortcut_is_accepted() {
+        let (tx_raw, rx) = channel::<AppEvent>();
+        let tx = AppEventSender::new(tx_raw);
+        let req = ApprovalRequest::Exec {
+            id: "1".to_string(),
+            command: vec!["echo".to_string()],
+            cwd: PathBuf::new(),
+            reason: None,
+        };
+        let mut widget = UserApprovalWidget::new(req, tx);
+        widget.handle_key_event(KeyEvent::new(KeyCode::Char('y'), KeyModifiers::NONE));
+        assert!(widget.is_complete());
+        let events: Vec<AppEvent> = rx.try_iter().collect();
+        assert!(events.iter().any(|e| matches!(
+            e,
+            AppEvent::CodexOp(Op::ExecApproval {
+                decision: ReviewDecision::Approved,
+                ..
+            })
+        )));
+    }
+
+    #[test]
+    fn uppercase_shortcut_is_accepted() {
+        let (tx_raw, rx) = channel::<AppEvent>();
+        let tx = AppEventSender::new(tx_raw);
+        let req = ApprovalRequest::Exec {
+            id: "2".to_string(),
+            command: vec!["echo".to_string()],
+            cwd: PathBuf::new(),
+            reason: None,
+        };
+        let mut widget = UserApprovalWidget::new(req, tx);
+        widget.handle_key_event(KeyEvent::new(KeyCode::Char('Y'), KeyModifiers::NONE));
+        assert!(widget.is_complete());
+        let events: Vec<AppEvent> = rx.try_iter().collect();
+        assert!(events.iter().any(|e| matches!(
+            e,
+            AppEvent::CodexOp(Op::ExecApproval {
+                decision: ReviewDecision::Approved,
+                ..
+            })
+        )));
+    }
+}