[apply-patch] Handle multiple context lines

## Summary
These tests were getting a bit unwieldy, and they're starting to become
load-bearing. Let's clean them up, and get them working solidly so we
can easily expand this harness with new tests.

## Test Plan
- [x] Tests continue to pass

codex-rs/Cargo.lock

@@ -635,6 +635,7 @@ name = "codex-apply-patch"
 version = "0.0.0"
 dependencies = [
  "anyhow",
+ "assert_cmd",
  "pretty_assertions",
  "similar",
  "tempfile",
@@ -652,6 +653,7 @@ dependencies = [
  "codex-core",
  "codex-linux-sandbox",
  "dotenvy",
+ "tempfile",
  "tokio",
 ]
 
@@ -731,6 +733,7 @@ dependencies = [
  "mime_guess",
  "openssl-sys",
  "os_info",
+ "portable-pty",
  "predicates",
  "pretty_assertions",
  "rand 0.9.2",
@@ -751,7 +754,7 @@ dependencies = [
  "tokio-test",
  "tokio-util",
  "toml 0.9.5",
- "toml_edit 0.23.3",
+ "toml_edit 0.23.4",
  "tracing",
  "tree-sitter",
  "tree-sitter-bash",
@@ -1479,6 +1482,12 @@ version = "0.15.7"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "1aaf95b3e5c8f23aa320147307562d361db0ae0d51242340f558153b4eb2439b"
 
+[[package]]
+name = "downcast-rs"
+version = "1.2.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "75b325c5dbd37f80359721ad39aca5a29fb04c89279657cffdda8736d0c0b9d2"
+
 [[package]]
 name = "dupe"
 version = "0.9.1"
@@ -1724,6 +1733,17 @@ dependencies = [
  "simd-adler32",
 ]
 
+[[package]]
+name = "filedescriptor"
+version = "0.8.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "e40758ed24c9b2eeb76c35fb0aebc66c626084edd827e07e1552279814c6682d"
+dependencies = [
+ "libc",
+ "thiserror 1.0.69",
+ "winapi",
+]
+
 [[package]]
 name = "fixedbitset"
 version = "0.4.2"
@@ -2702,6 +2722,7 @@ checksum = "4488594b9328dee448adb906d8b126d9b7deb7cf5c22161ee591610bb1be83c0"
 dependencies = [
  "bitflags 2.9.1",
  "libc",
+ "redox_syscall",
 ]
 
 [[package]]
@@ -3439,6 +3460,27 @@ dependencies = [
  "portable-atomic",
 ]
 
+[[package]]
+name = "portable-pty"
+version = "0.9.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "b4a596a2b3d2752d94f51fac2d4a96737b8705dddd311a32b9af47211f08671e"
+dependencies = [
+ "anyhow",
+ "bitflags 1.3.2",
+ "downcast-rs",
+ "filedescriptor",
+ "lazy_static",
+ "libc",
+ "log",
+ "nix",
+ "serial2",
+ "shared_library",
+ "shell-words",
+ "winapi",
+ "winreg",
+]
+
 [[package]]
 name = "potential_utf"
 version = "0.1.2"
@@ -4366,6 +4408,17 @@ dependencies = [
  "syn 2.0.104",
 ]
 
+[[package]]
+name = "serial2"
+version = "0.2.31"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "26e1e5956803a69ddd72ce2de337b577898801528749565def03515f82bad5bb"
+dependencies = [
+ "cfg-if",
+ "libc",
+ "winapi",
+]
+
 [[package]]
 name = "sha1"
 version = "0.10.6"
@@ -4397,6 +4450,22 @@ dependencies = [
  "lazy_static",
 ]
 
+[[package]]
+name = "shared_library"
+version = "0.1.9"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "5a9e7e0f2bfae24d8a5b5a66c5b257a83c7412304311512a0c054cd5e619da11"
+dependencies = [
+ "lazy_static",
+ "libc",
+]
+
+[[package]]
+name = "shell-words"
+version = "1.1.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "24188a676b6ae68c3b2cb3a01be17fbf7240ce009799bb56d5b1409051e78fde"
+
 [[package]]
 name = "shlex"
 version = "1.3.0"
@@ -5126,9 +5195,9 @@ dependencies = [
 
 [[package]]
 name = "toml_edit"
-version = "0.23.3"
+version = "0.23.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "17d3b47e6b7a040216ae5302712c94d1cf88c95b47efa80e2c59ce96c878267e"
+checksum = "7211ff1b8f0d3adae1663b7da9ffe396eabe1ca25f0b0bee42b0da29a9ddce93"
 dependencies = [
  "indexmap 2.10.0",
  "toml_datetime 0.7.0",
@@ -5709,11 +5778,11 @@ dependencies = [
 
 [[package]]
 name = "whoami"
-version = "1.6.0"
+version = "1.6.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "6994d13118ab492c3c80c1f81928718159254c53c472bf9ce36f8dae4add02a7"
+checksum = "5d4a4db5077702ca3015d3d02d74974948aba2ad9e12ab7df718ee64ccd7e97d"
 dependencies = [
- "redox_syscall",
+ "libredox",
  "wasite",
  "web-sys",
 ]
@@ -6176,6 +6245,15 @@ dependencies = [
  "memchr",
 ]
 
+[[package]]
+name = "winreg"
+version = "0.10.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "80d0f4e272c85def139476380b12f9ac60926689dd2e01d4923222f40580869d"
+dependencies = [
+ "winapi",
+]
+
 [[package]]
 name = "winsafe"
 version = "0.0.19"

codex-rs/README.md

@@ -43,6 +43,12 @@ To run Codex non-interactively, run `codex exec PROMPT` (you can also pass the p
 
 Typing `@` triggers a fuzzy-filename search over the workspace root. Use up/down to select among the results and Tab or Enter to replace the `@` with the selected path. You can use Esc to cancel the search.
 
+### Esc–Esc to edit a previous message
+
+When the chat composer is empty, press Esc to prime “backtrack” mode. Press Esc again to open a transcript preview highlighting the last user message; press Esc repeatedly to step to older user messages. Press Enter to confirm and Codex will fork the conversation from that point, trim the visible transcript accordingly, and pre‑fill the composer with the selected user message so you can edit and resubmit it.
+
+In the transcript preview, the footer shows an `Esc edit prev` hint while editing is active.
+
 ### `--cd`/`-C` flag
 
 Sometimes it is not convenient to `cd` to the directory you want Codex to use as the "working root" before running Codex. Fortunately, `codex` supports a `--cd` option so you can specify whatever folder you want. You can confirm that Codex is honoring `--cd` by double-checking the **workdir** it reports in the TUI at the start of a new session.

codex-rs/apply-patch/Cargo.toml

@@ -7,6 +7,10 @@ version = { workspace = true }
 name = "codex_apply_patch"
 path = "src/lib.rs"
 
+[[bin]]
+name = "apply_patch"
+path = "src/main.rs"
+
 [lints]
 workspace = true
 
@@ -18,5 +22,6 @@ tree-sitter = "0.25.8"
 tree-sitter-bash = "0.25.0"
 
 [dev-dependencies]
+assert_cmd = "2"
 pretty_assertions = "1.4.1"
 tempfile = "3.13.0"

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

@@ -1,5 +1,6 @@
 mod parser;
 mod seek_sequence;
+mod standalone_executable;
 
 use std::collections::HashMap;
 use std::path::Path;
@@ -19,6 +20,8 @@ use tree_sitter::LanguageError;
 use tree_sitter::Parser;
 use tree_sitter_bash::LANGUAGE as BASH;
 
+pub use standalone_executable::main;
+
 /// Detailed instructions for gpt-4.1 on how to use the `apply_patch` tool.
 pub const APPLY_PATCH_TOOL_INSTRUCTIONS: &str = include_str!("../apply_patch_tool_instructions.md");
 
@@ -529,32 +532,51 @@ fn compute_replacements(
     let mut line_index: usize = 0;
 
     for chunk in chunks {
-        // If a chunk has a `change_context`, we use seek_sequence to find it, then
-        // adjust our `line_index` to continue from there.
-        if let Some(ctx_line) = &chunk.change_context {
-            if let Some(idx) = seek_sequence::seek_sequence(
-                original_lines,
-                std::slice::from_ref(ctx_line),
-                line_index,
-                false,
-            ) {
-                line_index = idx + 1;
-            } else {
-                return Err(ApplyPatchError::ComputeReplacements(format!(
-                    "Failed to find context '{}' in {}",
-                    ctx_line,
-                    path.display()
-                )));
+        // If a chunk has context lines, we use seek_sequence to find each in order,
+        // then adjust our `line_index` to continue from there.
+        if !chunk.context_lines.is_empty() {
+            let total = chunk.context_lines.len();
+            for (i, ctx_line) in chunk.context_lines.iter().enumerate() {
+                if let Some(idx) = seek_sequence::seek_sequence(
+                    original_lines,
+                    std::slice::from_ref(ctx_line),
+                    line_index,
+                    false,
+                ) {
+                    line_index = idx + 1;
+                } else {
+                    return Err(ApplyPatchError::ComputeReplacements(format!(
+                        "Failed to find context {}/{}: '{}' in {}",
+                        i + 1,
+                        total,
+                        ctx_line,
+                        path.display()
+                    )));
+                }
             }
         }
 
         if chunk.old_lines.is_empty() {
-            // Pure addition (no old lines). We'll add them at the end or just
-            // before the final empty line if one exists.
-            let insertion_idx = if original_lines.last().is_some_and(|s| s.is_empty()) {
-                original_lines.len() - 1
+            // Pure addition (no old lines).
+            // Prefer to insert at the matched context anchor if one exists and
+            // the hunk is not explicitly marked as end-of-file.
+            let insertion_idx = if chunk.is_end_of_file {
+                if original_lines.last().is_some_and(|s| s.is_empty()) {
+                    original_lines.len() - 1
+                } else {
+                    original_lines.len()
+                }
+            } else if !chunk.context_lines.is_empty() {
+                // Insert immediately after the last matched context line.
+                line_index
             } else {
-                original_lines.len()
+                // No context provided: fall back to appending at the end (before
+                // the trailing empty line if present).
+                if original_lines.last().is_some_and(|s| s.is_empty()) {
+                    original_lines.len() - 1
+                } else {
+                    original_lines.len()
+                }
             };
             replacements.push((insertion_idx, 0, chunk.new_lines.clone()));
             continue;
@@ -1267,6 +1289,57 @@ g
         );
     }
 
+    #[test]
+    fn test_insert_addition_after_single_context_anchor() {
+        let dir = tempdir().unwrap();
+        let path = dir.path().join("single_ctx.txt");
+        fs::write(&path, "class BaseClass:\n    def method():\nline1\nline2\n").unwrap();
+
+        let patch = wrap_patch(&format!(
+            r#"*** Update File: {}
+@@ class BaseClass:
++INSERTED
+"#,
+            path.display()
+        ));
+
+        let mut stdout = Vec::new();
+        let mut stderr = Vec::new();
+        apply_patch(&patch, &mut stdout, &mut stderr).unwrap();
+
+        let contents = fs::read_to_string(path).unwrap();
+        assert_eq!(
+            contents,
+            "class BaseClass:\nINSERTED\n    def method():\nline1\nline2\n"
+        );
+    }
+
+    #[test]
+    fn test_insert_addition_after_multi_context_anchor() {
+        let dir = tempdir().unwrap();
+        let path = dir.path().join("multi_ctx.txt");
+        fs::write(&path, "class BaseClass:\n    def method():\nline1\nline2\n").unwrap();
+
+        let patch = wrap_patch(&format!(
+            r#"*** Update File: {}
+@@ class BaseClass:
+@@     def method():
++INSERTED
+"#,
+            path.display()
+        ));
+
+        let mut stdout = Vec::new();
+        let mut stderr = Vec::new();
+        apply_patch(&patch, &mut stdout, &mut stderr).unwrap();
+
+        let contents = fs::read_to_string(path).unwrap();
+        assert_eq!(
+            contents,
+            "class BaseClass:\n    def method():\nINSERTED\nline1\nline2\n"
+        );
+    }
+
     #[test]
     fn test_apply_patch_should_resolve_absolute_paths_in_cwd() {
         let session_dir = tempdir().unwrap();

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

@@ -0,0 +1,3 @@
+pub fn main() -> ! {
+    codex_apply_patch::main()
+}

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

@@ -69,7 +69,7 @@ pub enum Hunk {
         path: PathBuf,
         move_path: Option<PathBuf>,
 
-        /// Chunks should be in order, i.e. the `change_context` of one chunk
+        /// Chunks should be in order, i.e. the first context line of one chunk
         /// should occur later in the file than the previous chunk.
         chunks: Vec<UpdateFileChunk>,
     },
@@ -89,12 +89,13 @@ use Hunk::*;
 
 #[derive(Debug, PartialEq, Clone)]
 pub struct UpdateFileChunk {
-    /// A single line of context used to narrow down the position of the chunk
-    /// (this is usually a class, method, or function definition.)
-    pub change_context: Option<String>,
+    /// Context lines used to narrow down the position of the chunk.
+    /// Each entry is searched sequentially to progressively restrict the
+    /// search to the desired region (e.g. class → method).
+    pub context_lines: Vec<String>,
 
     /// A contiguous block of lines that should be replaced with `new_lines`.
-    /// `old_lines` must occur strictly after `change_context`.
+    /// `old_lines` must occur strictly after the context.
     pub old_lines: Vec<String>,
     pub new_lines: Vec<String>,
 
@@ -344,32 +345,38 @@ fn parse_update_file_chunk(
             line_number,
         });
     }
-    // If we see an explicit context marker @@ or @@ <context>, consume it; otherwise, optionally
-    // allow treating the chunk as starting directly with diff lines.
-    let (change_context, start_index) = if lines[0] == EMPTY_CHANGE_CONTEXT_MARKER {
-        (None, 1)
-    } else if let Some(context) = lines[0].strip_prefix(CHANGE_CONTEXT_MARKER) {
-        (Some(context.to_string()), 1)
-    } else {
-        if !allow_missing_context {
-            return Err(InvalidHunkError {
-                message: format!(
-                    "Expected update hunk to start with a @@ context marker, got: '{}'",
-                    lines[0]
-                ),
-                line_number,
-            });
+    let mut context_lines = Vec::new();
+    let mut start_index = 0;
+    let mut saw_context_marker = false;
+    while start_index < lines.len() {
+        if lines[start_index] == EMPTY_CHANGE_CONTEXT_MARKER {
+            saw_context_marker = true;
+            start_index += 1;
+        } else if let Some(context) = lines[start_index].strip_prefix(CHANGE_CONTEXT_MARKER) {
+            saw_context_marker = true;
+            context_lines.push(context.to_string());
+            start_index += 1;
+        } else {
+            break;
         }
-        (None, 0)
-    };
+    }
+    if !saw_context_marker && !allow_missing_context {
+        return Err(InvalidHunkError {
+            message: format!(
+                "Expected update hunk to start with a @@ context marker, got: '{}'",
+                lines[0]
+            ),
+            line_number,
+        });
+    }
     if start_index >= lines.len() {
         return Err(InvalidHunkError {
             message: "Update hunk does not contain any lines".to_string(),
-            line_number: line_number + 1,
+            line_number: line_number + start_index,
         });
     }
     let mut chunk = UpdateFileChunk {
-        change_context,
+        context_lines,
         old_lines: Vec::new(),
         new_lines: Vec::new(),
         is_end_of_file: false,
@@ -381,7 +388,7 @@ fn parse_update_file_chunk(
                 if parsed_lines == 0 {
                     return Err(InvalidHunkError {
                         message: "Update hunk does not contain any lines".to_string(),
-                        line_number: line_number + 1,
+                        line_number: line_number + start_index,
                     });
                 }
                 chunk.is_end_of_file = true;
@@ -411,7 +418,7 @@ fn parse_update_file_chunk(
                                 message: format!(
                                     "Unexpected line found in update hunk: '{line_contents}'. Every line should start with ' ' (context line), '+' (added line), or '-' (removed line)"
                                 ),
-                                line_number: line_number + 1,
+                                line_number: line_number + start_index,
                             });
                         }
                         // Assume this is the start of the next hunk.
@@ -491,7 +498,7 @@ fn test_parse_patch() {
                 path: PathBuf::from("path/update.py"),
                 move_path: Some(PathBuf::from("path/update2.py")),
                 chunks: vec![UpdateFileChunk {
-                    change_context: Some("def f():".to_string()),
+                    context_lines: vec!["def f():".to_string()],
                     old_lines: vec!["    pass".to_string()],
                     new_lines: vec!["    return 123".to_string()],
                     is_end_of_file: false
@@ -518,7 +525,7 @@ fn test_parse_patch() {
                 path: PathBuf::from("file.py"),
                 move_path: None,
                 chunks: vec![UpdateFileChunk {
-                    change_context: None,
+                    context_lines: Vec::new(),
                     old_lines: vec![],
                     new_lines: vec!["line".to_string()],
                     is_end_of_file: false
@@ -548,7 +555,7 @@ fn test_parse_patch() {
             path: PathBuf::from("file2.py"),
             move_path: None,
             chunks: vec![UpdateFileChunk {
-                change_context: None,
+                context_lines: Vec::new(),
                 old_lines: vec!["import foo".to_string()],
                 new_lines: vec!["import foo".to_string(), "bar".to_string()],
                 is_end_of_file: false,
@@ -568,7 +575,7 @@ fn test_parse_patch_lenient() {
         path: PathBuf::from("file2.py"),
         move_path: None,
         chunks: vec![UpdateFileChunk {
-            change_context: None,
+            context_lines: Vec::new(),
             old_lines: vec!["import foo".to_string()],
             new_lines: vec!["import foo".to_string(), "bar".to_string()],
             is_end_of_file: false,
@@ -701,7 +708,7 @@ fn test_update_file_chunk() {
         ),
         Ok((
             (UpdateFileChunk {
-                change_context: Some("change_context".to_string()),
+                context_lines: vec!["change_context".to_string()],
                 old_lines: vec![
                     "".to_string(),
                     "context".to_string(),
@@ -723,7 +730,7 @@ fn test_update_file_chunk() {
         parse_update_file_chunk(&["@@", "+line", "*** End of File"], 123, false),
         Ok((
             (UpdateFileChunk {
-                change_context: None,
+                context_lines: Vec::new(),
                 old_lines: vec![],
                 new_lines: vec!["line".to_string()],
                 is_end_of_file: true
@@ -731,4 +738,29 @@ fn test_update_file_chunk() {
             3
         ))
     );
+    assert_eq!(
+        parse_update_file_chunk(
+            &[
+                "@@ class BaseClass",
+                "@@     def method()",
+                " context",
+                "-old",
+                "+new",
+            ],
+            123,
+            false
+        ),
+        Ok((
+            (UpdateFileChunk {
+                context_lines: vec![
+                    "class BaseClass".to_string(),
+                    "    def method()".to_string()
+                ],
+                old_lines: vec!["context".to_string(), "old".to_string()],
+                new_lines: vec!["context".to_string(), "new".to_string()],
+                is_end_of_file: false
+            }),
+            5
+        ))
+    );
 }

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

@@ -0,0 +1,59 @@
+use std::io::Read;
+use std::io::Write;
+
+pub fn main() -> ! {
+    let exit_code = run_main();
+    std::process::exit(exit_code);
+}
+
+/// We would prefer to return `std::process::ExitCode`, but its `exit_process()`
+/// method is still a nightly API and we want main() to return !.
+pub fn run_main() -> i32 {
+    // Expect either one argument (the full apply_patch payload) or read it from stdin.
+    let mut args = std::env::args_os();
+    let _argv0 = args.next();
+
+    let patch_arg = match args.next() {
+        Some(arg) => match arg.into_string() {
+            Ok(s) => s,
+            Err(_) => {
+                eprintln!("Error: apply_patch requires a UTF-8 PATCH argument.");
+                return 1;
+            }
+        },
+        None => {
+            // No argument provided; attempt to read the patch from stdin.
+            let mut buf = String::new();
+            match std::io::stdin().read_to_string(&mut buf) {
+                Ok(_) => {
+                    if buf.is_empty() {
+                        eprintln!("Usage: apply_patch 'PATCH'\n       echo 'PATCH' | apply-patch");
+                        return 2;
+                    }
+                    buf
+                }
+                Err(err) => {
+                    eprintln!("Error: Failed to read PATCH from stdin.\n{err}");
+                    return 1;
+                }
+            }
+        }
+    };
+
+    // Refuse extra args to avoid ambiguity.
+    if args.next().is_some() {
+        eprintln!("Error: apply_patch accepts exactly one argument.");
+        return 2;
+    }
+
+    let mut stdout = std::io::stdout();
+    let mut stderr = std::io::stderr();
+    match crate::apply_patch(&patch_arg, &mut stdout, &mut stderr) {
+        Ok(()) => {
+            // Flush to ensure output ordering when used in pipelines.
+            let _ = stdout.flush();
+            0
+        }
+        Err(_) => 1,
+    }
+}

codex-rs/apply-patch/tests/all.rs

@@ -0,0 +1,3 @@
+// Single integration test binary that aggregates all test modules.
+// The submodules live in `tests/suite/`.
+mod suite;

codex-rs/apply-patch/tests/suite/cli.rs

@@ -0,0 +1,90 @@
+use assert_cmd::prelude::*;
+use std::fs;
+use std::process::Command;
+use tempfile::tempdir;
+
+#[test]
+fn test_apply_patch_cli_add_and_update() -> anyhow::Result<()> {
+    let tmp = tempdir()?;
+    let file = "cli_test.txt";
+    let absolute_path = tmp.path().join(file);
+
+    // 1) Add a file
+    let add_patch = format!(
+        r#"*** Begin Patch
+*** Add File: {file}
++hello
+*** End Patch"#
+    );
+    Command::cargo_bin("apply_patch")
+        .expect("should find apply_patch binary")
+        .arg(add_patch)
+        .current_dir(tmp.path())
+        .assert()
+        .success()
+        .stdout(format!("Success. Updated the following files:\nA {file}\n"));
+    assert_eq!(fs::read_to_string(&absolute_path)?, "hello\n");
+
+    // 2) Update the file
+    let update_patch = format!(
+        r#"*** Begin Patch
+*** Update File: {file}
+@@
+-hello
++world
+*** End Patch"#
+    );
+    Command::cargo_bin("apply_patch")
+        .expect("should find apply_patch binary")
+        .arg(update_patch)
+        .current_dir(tmp.path())
+        .assert()
+        .success()
+        .stdout(format!("Success. Updated the following files:\nM {file}\n"));
+    assert_eq!(fs::read_to_string(&absolute_path)?, "world\n");
+
+    Ok(())
+}
+
+#[test]
+fn test_apply_patch_cli_stdin_add_and_update() -> anyhow::Result<()> {
+    let tmp = tempdir()?;
+    let file = "cli_test_stdin.txt";
+    let absolute_path = tmp.path().join(file);
+
+    // 1) Add a file via stdin
+    let add_patch = format!(
+        r#"*** Begin Patch
+*** Add File: {file}
++hello
+*** End Patch"#
+    );
+    let mut cmd =
+        assert_cmd::Command::cargo_bin("apply_patch").expect("should find apply_patch binary");
+    cmd.current_dir(tmp.path());
+    cmd.write_stdin(add_patch)
+        .assert()
+        .success()
+        .stdout(format!("Success. Updated the following files:\nA {file}\n"));
+    assert_eq!(fs::read_to_string(&absolute_path)?, "hello\n");
+
+    // 2) Update the file via stdin
+    let update_patch = format!(
+        r#"*** Begin Patch
+*** Update File: {file}
+@@
+-hello
++world
+*** End Patch"#
+    );
+    let mut cmd =
+        assert_cmd::Command::cargo_bin("apply_patch").expect("should find apply_patch binary");
+    cmd.current_dir(tmp.path());
+    cmd.write_stdin(update_patch)
+        .assert()
+        .success()
+        .stdout(format!("Success. Updated the following files:\nM {file}\n"));
+    assert_eq!(fs::read_to_string(&absolute_path)?, "world\n");
+
+    Ok(())
+}

codex-rs/apply-patch/tests/suite/mod.rs

@@ -0,0 +1 @@
+mod cli;

codex-rs/arg0/Cargo.toml

@@ -16,4 +16,5 @@ codex-apply-patch = { path = "../apply-patch" }
 codex-core = { path = "../core" }
 codex-linux-sandbox = { path = "../linux-sandbox" }
 dotenvy = "0.15.7"
+tempfile = "3"
 tokio = { version = "1", features = ["rt-multi-thread"] }

codex-rs/arg0/src/lib.rs

@@ -3,6 +3,13 @@ use std::path::Path;
 use std::path::PathBuf;
 
 use codex_core::CODEX_APPLY_PATCH_ARG1;
+#[cfg(unix)]
+use std::os::unix::fs::symlink;
+use tempfile::TempDir;
+
+const LINUX_SANDBOX_ARG0: &str = "codex-linux-sandbox";
+const APPLY_PATCH_ARG0: &str = "apply_patch";
+const MISSPELLED_APPLY_PATCH_ARG0: &str = "applypatch";
 
 /// While we want to deploy the Codex CLI as a single executable for simplicity,
 /// we also want to expose some of its functionality as distinct CLIs, so we use
@@ -39,9 +46,11 @@ where
         .and_then(|s| s.to_str())
         .unwrap_or("");
 
-    if exe_name == "codex-linux-sandbox" {
+    if exe_name == LINUX_SANDBOX_ARG0 {
         // Safety: [`run_main`] never returns.
         codex_linux_sandbox::run_main();
+    } else if exe_name == APPLY_PATCH_ARG0 || exe_name == MISSPELLED_APPLY_PATCH_ARG0 {
+        codex_apply_patch::main();
     }
 
     let argv1 = args.next().unwrap_or_default();
@@ -68,6 +77,19 @@ where
     // before creating any threads/the Tokio runtime.
     load_dotenv();
 
+    // Retain the TempDir so it exists for the lifetime of the invocation of
+    // this executable. Admittedly, we could invoke `keep()` on it, but it
+    // would be nice to avoid leaving temporary directories behind, if possible.
+    let _path_entry = match prepend_path_entry_for_apply_patch() {
+        Ok(path_entry) => Some(path_entry),
+        Err(err) => {
+            // It is possible that Codex will proceed successfully even if
+            // updating the PATH fails, so warn the user and move on.
+            eprintln!("WARNING: proceeding, even though we could not update PATH: {err}");
+            None
+        }
+    };
+
     // Regular invocation – create a Tokio runtime and execute the provided
     // async entry-point.
     let runtime = tokio::runtime::Runtime::new()?;
@@ -113,3 +135,67 @@ where
         }
     }
 }
+
+/// Creates a temporary directory with either:
+///
+/// - UNIX: `apply_patch` symlink to the current executable
+/// - WINDOWS: `apply_patch.bat` batch script to invoke the current executable
+///   with the "secret" --codex-run-as-apply-patch flag.
+///
+/// This temporary directory is prepended to the PATH environment variable so
+/// that `apply_patch` can be on the PATH without requiring the user to
+/// install a separate `apply_patch` executable, simplifying the deployment of
+/// Codex CLI.
+///
+/// IMPORTANT: This function modifies the PATH environment variable, so it MUST
+/// be called before multiple threads are spawned.
+fn prepend_path_entry_for_apply_patch() -> std::io::Result<TempDir> {
+    let temp_dir = TempDir::new()?;
+    let path = temp_dir.path();
+
+    for filename in &[APPLY_PATCH_ARG0, MISSPELLED_APPLY_PATCH_ARG0] {
+        let exe = std::env::current_exe()?;
+
+        #[cfg(unix)]
+        {
+            let link = path.join(filename);
+            symlink(&exe, &link)?;
+        }
+
+        #[cfg(windows)]
+        {
+            let batch_script = path.join(format!("{filename}.bat"));
+            std::fs::write(
+                &batch_script,
+                format!(
+                    r#"@echo off
+"{}" {CODEX_APPLY_PATCH_ARG1} %*
+"#,
+                    exe.display()
+                ),
+            )?;
+        }
+    }
+
+    #[cfg(unix)]
+    const PATH_SEPARATOR: &str = ":";
+
+    #[cfg(windows)]
+    const PATH_SEPARATOR: &str = ";";
+
+    let path_element = path.display();
+    let updated_path_env_var = match std::env::var("PATH") {
+        Ok(existing_path) => {
+            format!("{path_element}{PATH_SEPARATOR}{existing_path}")
+        }
+        Err(_) => {
+            format!("{path_element}")
+        }
+    };
+
+    unsafe {
+        std::env::set_var("PATH", updated_path_env_var);
+    }
+
+    Ok(temp_dir)
+}

codex-rs/chatgpt/tests/all.rs

@@ -0,0 +1,3 @@
+// Single integration test binary that aggregates all test modules.
+// The submodules live in `tests/suite/`.
+mod suite;

codex-rs/chatgpt/tests/suite/mod.rs

@@ -0,0 +1,2 @@
+// Aggregates all former standalone integration tests as modules.
+mod apply_command_e2e;

codex-rs/core/Cargo.toml

@@ -6,6 +6,7 @@ version = { workspace = true }
 [lib]
 name = "codex_core"
 path = "src/lib.rs"
+doctest = false
 
 [lints]
 workspace = true
@@ -28,6 +29,7 @@ libc = "0.2.175"
 mcp-types = { path = "../mcp-types" }
 mime_guess = "2.0"
 os_info = "3.12.0"
+portable-pty = "0.9.0"
 rand = "0.9"
 regex-lite = "0.1.6"
 reqwest = { version = "0.12", features = ["json", "stream"] }
@@ -50,12 +52,12 @@ tokio = { version = "1", features = [
 ] }
 tokio-util = "0.7.16"
 toml = "0.9.5"
-toml_edit = "0.23.3"
+toml_edit = "0.23.4"
 tracing = { version = "0.1.41", features = ["log"] }
 tree-sitter = "0.25.8"
 tree-sitter-bash = "0.25.0"
 uuid = { version = "1", features = ["serde", "v4"] }
-whoami = "1.6.0"
+whoami = "1.6.1"
 wildmatch = "2.4.0"
 
 

codex-rs/core/src/chat_completions.rs

@@ -623,6 +623,12 @@ where
                 Poll::Ready(Some(Ok(ResponseEvent::ReasoningSummaryPartAdded))) => {
                     continue;
                 }
+                Poll::Ready(Some(Ok(ResponseEvent::WebSearchCallBegin { .. }))) => {
+                    return Poll::Ready(Some(Ok(ResponseEvent::WebSearchCallBegin {
+                        call_id: String::new(),
+                        query: None,
+                    })));
+                }
             }
         }
     }

codex-rs/core/src/client.rs

@@ -149,7 +149,21 @@ impl ModelClient {
         let store = prompt.store && auth_mode != Some(AuthMode::ChatGPT);
 
         let full_instructions = prompt.get_full_instructions(&self.config.model_family);
-        let tools_json = create_tools_json_for_responses_api(&prompt.tools)?;
+        let mut tools_json = create_tools_json_for_responses_api(&prompt.tools)?;
+        // ChatGPT backend expects the preview name for web search.
+        if auth_mode == Some(AuthMode::ChatGPT) {
+            for tool in &mut tools_json {
+                if let Some(map) = tool.as_object_mut()
+                    && map.get("type").and_then(|v| v.as_str()) == Some("web_search")
+                {
+                    map.insert(
+                        "type".to_string(),
+                        serde_json::Value::String("web_search_preview".to_string()),
+                    );
+                }
+            }
+        }
+
         let reasoning = create_reasoning_param_for_request(
             &self.config.model_family,
             self.effort,
@@ -466,7 +480,8 @@ async fn process_sse<S>(
             }
         };
 
-        trace!("SSE event: {}", sse.data);
+        let raw = sse.data.clone();
+        trace!("SSE event: {}", raw);
 
         let event: SseEvent = match serde_json::from_str(&sse.data) {
             Ok(event) => event,
@@ -580,8 +595,24 @@ async fn process_sse<S>(
             | "response.in_progress"
             | "response.output_item.added"
             | "response.output_text.done" => {
-                // Currently, we ignore this event, but we handle it
-                // separately to skip the logging message in the `other` case.
+                if event.kind == "response.output_item.added"
+                    && let Some(item) = event.item.as_ref()
+                {
+                    // Detect web_search_call begin and forward a synthetic event upstream.
+                    if let Some(ty) = item.get("type").and_then(|v| v.as_str())
+                        && ty == "web_search_call"
+                    {
+                        let call_id = item
+                            .get("id")
+                            .and_then(|v| v.as_str())
+                            .unwrap_or("")
+                            .to_string();
+                        let ev = ResponseEvent::WebSearchCallBegin { call_id, query: None };
+                        if tx_event.send(Ok(ev)).await.is_err() {
+                            return;
+                        }
+                    }
+                }
             }
             "response.reasoning_summary_part.added" => {
                 // Boundary between reasoning summary sections (e.g., titles).
@@ -591,7 +622,7 @@ async fn process_sse<S>(
                 }
             }
             "response.reasoning_summary_text.done" => {}
-            other => debug!(other, "sse event"),
+            _ => {}
         }
     }
 }

codex-rs/core/src/client_common.rs

@@ -93,6 +93,10 @@ pub enum ResponseEvent {
     ReasoningSummaryDelta(String),
     ReasoningContentDelta(String),
     ReasoningSummaryPartAdded,
+    WebSearchCallBegin {
+        call_id: String,
+        query: Option<String>,
+    },
 }
 
 #[derive(Debug, Serialize)]

codex-rs/core/src/codex.rs

@@ -53,12 +53,18 @@ use crate::exec::SandboxType;
 use crate::exec::StdoutStream;
 use crate::exec::StreamOutput;
 use crate::exec::process_exec_tool_call;
+use crate::exec_command::EXEC_COMMAND_TOOL_NAME;
+use crate::exec_command::ExecCommandParams;
+use crate::exec_command::ExecSessionManager;
+use crate::exec_command::WRITE_STDIN_TOOL_NAME;
+use crate::exec_command::WriteStdinParams;
 use crate::exec_env::create_env;
 use crate::mcp_connection_manager::McpConnectionManager;
 use crate::mcp_tool_call::handle_mcp_tool_call;
 use crate::model_family::find_family_for_model;
 use crate::openai_tools::ApplyPatchToolArgs;
 use crate::openai_tools::ToolsConfig;
+use crate::openai_tools::ToolsConfigParams;
 use crate::openai_tools::get_openai_tools;
 use crate::parse_command::parse_command;
 use crate::plan_tool::handle_update_plan;
@@ -91,6 +97,7 @@ use crate::protocol::StreamErrorEvent;
 use crate::protocol::Submission;
 use crate::protocol::TaskCompleteEvent;
 use crate::protocol::TurnDiffEvent;
+use crate::protocol::WebSearchBeginEvent;
 use crate::rollout::RolloutRecorder;
 use crate::safety::SafetyCheck;
 use crate::safety::assess_command_safety;
@@ -262,15 +269,7 @@ pub(crate) struct Session {
 
     /// Manager for external MCP servers/tools.
     mcp_connection_manager: McpConnectionManager,
-
-    /// Loaded subagent definitions from project and user scope.
-    subagents_registry: crate::subagents::registry::SubagentRegistry,
-
-    /// Auth manager used to spawn nested sessions (e.g., subagents).
-    auth_manager: Arc<AuthManager>,
-
-    /// Base configuration used to derive nested session configs.
-    base_config: Arc<Config>,
+    session_manager: ExecSessionManager,
 
     /// External notifier command (will be passed as args to exec()). When
     /// `None` this feature is disabled.
@@ -507,40 +506,17 @@ impl Session {
             model_reasoning_summary,
             session_id,
         );
-        // Build subagent registry paths and load once per session
-        let project_agents_dir = {
-            let mut p = cwd.clone();
-            p.push(".codex");
-            p.push("agents");
-            if p.exists() { Some(p) } else { None }
-        };
-        let user_agents_dir = {
-            let mut p = config.codex_home.clone();
-            p.push("agents");
-            if p.exists() { Some(p) } else { None }
-        };
-        let mut subagents_registry =
-            crate::subagents::registry::SubagentRegistry::new(project_agents_dir, user_agents_dir);
-        subagents_registry.load();
-        // Log discovered subagents for visibility in clients (e.g., TUI) after
-        // SessionConfigured so the first event contract remains intact.
-        post_session_configured_error_events.push(Event {
-            id: INITIAL_SUBMIT_ID.to_string(),
-            msg: EventMsg::BackgroundEvent(BackgroundEventEvent {
-                message: format!("subagents discovered: {:?}", subagents_registry.all_names()),
-            }),
-        });
-
         let turn_context = TurnContext {
             client,
-            tools_config: ToolsConfig::new(
-                &config.model_family,
+            tools_config: ToolsConfig::new(&ToolsConfigParams {
+                model_family: &config.model_family,
                 approval_policy,
-                sandbox_policy.clone(),
-                config.include_plan_tool,
-                config.include_apply_patch_tool,
-                config.include_subagent_tool,
-            ),
+                sandbox_policy: sandbox_policy.clone(),
+                include_plan_tool: config.include_plan_tool,
+                include_apply_patch_tool: config.include_apply_patch_tool,
+                include_web_search_request: config.tools_web_search_request,
+                use_streamable_shell_tool: config.use_experimental_streamable_shell_tool,
+            }),
             user_instructions,
             base_instructions,
             approval_policy,
@@ -553,9 +529,7 @@ impl Session {
             session_id,
             tx_event: tx_event.clone(),
             mcp_connection_manager,
-            subagents_registry,
-            auth_manager: auth_manager.clone(),
-            base_config: config.clone(),
+            session_manager: ExecSessionManager::default(),
             notify,
             state: Mutex::new(state),
             rollout: Mutex::new(rollout_recorder),
@@ -615,16 +589,6 @@ impl Session {
         }
     }
 
-    /// Access auth manager for nested sessions.
-    pub(crate) fn auth_manager(&self) -> Arc<AuthManager> {
-        self.auth_manager.clone()
-    }
-
-    /// Access base config for nested sessions.
-    pub(crate) fn base_config(&self) -> Arc<Config> {
-        self.base_config.clone()
-    }
-
     /// Sends the given event to the client and swallows the send event, if
     /// any, logging it as an error.
     pub(crate) async fn send_event(&self, event: Event) {
@@ -1131,14 +1095,15 @@ async fn submission_loop(
                     .unwrap_or(prev.sandbox_policy.clone());
                 let new_cwd = cwd.clone().unwrap_or_else(|| prev.cwd.clone());
 
-                let tools_config = ToolsConfig::new(
-                    &effective_family,
-                    new_approval_policy,
-                    new_sandbox_policy.clone(),
-                    config.include_plan_tool,
-                    config.include_apply_patch_tool,
-                    config.include_subagent_tool,
-                );
+                let tools_config = ToolsConfig::new(&ToolsConfigParams {
+                    model_family: &effective_family,
+                    approval_policy: new_approval_policy,
+                    sandbox_policy: new_sandbox_policy.clone(),
+                    include_plan_tool: config.include_plan_tool,
+                    include_apply_patch_tool: config.include_apply_patch_tool,
+                    include_web_search_request: config.tools_web_search_request,
+                    use_streamable_shell_tool: config.use_experimental_streamable_shell_tool,
+                });
 
                 let new_turn_context = TurnContext {
                     client,
@@ -1210,14 +1175,16 @@ async fn submission_loop(
 
                     let fresh_turn_context = TurnContext {
                         client,
-                        tools_config: ToolsConfig::new(
-                            &model_family,
+                        tools_config: ToolsConfig::new(&ToolsConfigParams {
+                            model_family: &model_family,
                             approval_policy,
-                            sandbox_policy.clone(),
-                            config.include_plan_tool,
-                            config.include_apply_patch_tool,
-                            config.include_subagent_tool,
-                        ),
+                            sandbox_policy: sandbox_policy.clone(),
+                            include_plan_tool: config.include_plan_tool,
+                            include_apply_patch_tool: config.include_apply_patch_tool,
+                            include_web_search_request: config.tools_web_search_request,
+                            use_streamable_shell_tool: config
+                                .use_experimental_streamable_shell_tool,
+                        }),
                         user_instructions: turn_context.user_instructions.clone(),
                         base_instructions: turn_context.base_instructions.clone(),
                         approval_policy,
@@ -1603,27 +1570,6 @@ async fn run_turn(
         &turn_context.tools_config,
         Some(sess.mcp_connection_manager.list_all_tools()),
     );
-    tracing::trace!("Tools: {tools:?}");
-
-    // Log tool names for visibility in the TUI/debug logs.
-    #[allow(clippy::match_same_arms)]
-    let tool_names: Vec<String> = tools
-        .iter()
-        .map(|t| match t {
-            crate::openai_tools::OpenAiTool::Function(f) => f.name.clone(),
-            crate::openai_tools::OpenAiTool::LocalShell {} => "local_shell".to_string(),
-            crate::openai_tools::OpenAiTool::Freeform(f) => f.name.clone(),
-        })
-        .collect();
-    let _ = sess
-        .tx_event
-        .send(Event {
-            id: sub_id.clone(),
-            msg: EventMsg::BackgroundEvent(BackgroundEventEvent {
-                message: format!("tools available: {:?}", tool_names),
-            }),
-        })
-        .await;
 
     let prompt = Prompt {
         input,
@@ -1749,6 +1695,7 @@ async fn try_run_turn(
     let mut stream = turn_context.client.clone().stream(&prompt).await?;
 
     let mut output = Vec::new();
+
     loop {
         // Poll the next item from the model stream. We must inspect *both* Ok and Err
         // cases so that transient stream failures (e.g., dropped SSE connection before
@@ -1785,6 +1732,16 @@ async fn try_run_turn(
                 .await?;
                 output.push(ProcessedResponseItem { item, response });
             }
+            ResponseEvent::WebSearchCallBegin { call_id, query } => {
+                let q = query.unwrap_or_else(|| "Searching Web...".to_string());
+                let _ = sess
+                    .tx_event
+                    .send(Event {
+                        id: sub_id.to_string(),
+                        msg: EventMsg::WebSearchBegin(WebSearchBeginEvent { call_id, query: q }),
+                    })
+                    .await;
+            }
             ResponseEvent::Completed {
                 response_id: _,
                 token_usage,
@@ -2143,16 +2100,10 @@ async fn handle_function_call(
             .await
         }
         "update_plan" => handle_update_plan(sess, arguments, sub_id, call_id).await,
-        "subagent_run" => {
-            #[derive(serde::Deserialize)]
-            struct Args {
-                name: String,
-                input: String,
-                #[serde(default)]
-                context: Option<String>,
-            }
-            let args = match serde_json::from_str::<Args>(&arguments) {
-                Ok(a) => a,
+        EXEC_COMMAND_TOOL_NAME => {
+            // TODO(mbolin): Sandbox check.
+            let exec_params = match serde_json::from_str::<ExecCommandParams>(&arguments) {
+                Ok(params) => params,
                 Err(e) => {
                     return ResponseInputItem::FunctionCallOutput {
                         call_id,
@@ -2163,62 +2114,38 @@ async fn handle_function_call(
                     };
                 }
             };
-
-            let result = crate::subagents::runner::run(
-                sess,
-                turn_context,
-                &sess.subagents_registry,
-                crate::subagents::runner::RunSubagentArgs {
-                    name: args.name,
-                    input: args.input,
-                    context: args.context,
-                },
-                &sub_id,
-            )
-            .await;
-
-            match result {
-                Ok(message) => ResponseInputItem::FunctionCallOutput {
-                    call_id,
-                    output: FunctionCallOutputPayload {
-                        content: message,
-                        success: Some(true),
-                    },
-                },
-                Err(e) => ResponseInputItem::FunctionCallOutput {
-                    call_id,
-                    output: FunctionCallOutputPayload {
-                        content: format!("subagent failed: {e}"),
-                        success: Some(false),
-                    },
-                },
-            }
-        }
-        "subagent_list" => {
-            #[derive(serde::Serialize)]
-            struct SubagentBrief<'a> {
-                name: &'a str,
-                description: &'a str,
-            }
-            let mut list = Vec::new();
-            for name in sess.subagents_registry.all_names() {
-                if let Some(def) = sess.subagents_registry.get(&name) {
-                    list.push(SubagentBrief {
-                        name: &def.name,
-                        description: &def.description,
-                    });
-                }
-            }
-            let payload = match serde_json::to_string(&list) {
-                Ok(s) => s,
-                Err(e) => format!("failed to serialize subagent list: {e}"),
-            };
+            let result = sess
+                .session_manager
+                .handle_exec_command_request(exec_params)
+                .await;
+            let function_call_output = crate::exec_command::result_into_payload(result);
             ResponseInputItem::FunctionCallOutput {
                 call_id,
-                output: FunctionCallOutputPayload {
-                    content: payload,
-                    success: Some(true),
-                },
+                output: function_call_output,
+            }
+        }
+        WRITE_STDIN_TOOL_NAME => {
+            let write_stdin_params = match serde_json::from_str::<WriteStdinParams>(&arguments) {
+                Ok(params) => params,
+                Err(e) => {
+                    return ResponseInputItem::FunctionCallOutput {
+                        call_id,
+                        output: FunctionCallOutputPayload {
+                            content: format!("failed to parse function arguments: {e}"),
+                            success: Some(false),
+                        },
+                    };
+                }
+            };
+            let result = sess
+                .session_manager
+                .handle_write_stdin_request(write_stdin_params)
+                .await;
+            let function_call_output: FunctionCallOutputPayload =
+                crate::exec_command::result_into_payload(result);
+            ResponseInputItem::FunctionCallOutput {
+                call_id,
+                output: function_call_output,
             }
         }
         _ => {
@@ -2331,8 +2258,6 @@ fn parse_container_exec_arguments(
     }
 }
 
-// (helper run_one_turn_collect removed as unused)
-
 pub struct ExecInvokeArgs<'a> {
     pub params: ExecParams,
     pub sandbox_type: SandboxType,
@@ -2891,15 +2816,9 @@ async fn drain_to_completed(
                 response_id: _,
                 token_usage,
             }) => {
-                let token_usage = match token_usage {
-                    Some(usage) => usage,
-                    None => {
-                        return Err(CodexErr::Stream(
-                            "token_usage was None in ResponseEvent::Completed".into(),
-                            None,
-                        ));
-                    }
-                };
+                // some providers don't return token usage, so we default
+                // TODO: consider approximate token usage
+                let token_usage = token_usage.unwrap_or_default();
                 sess.tx_event
                     .send(Event {
                         id: sub_id.to_string(),
@@ -2907,6 +2826,7 @@ async fn drain_to_completed(
                     })
                     .await
                     .ok();
+
                 return Ok(());
             }
             Ok(_) => continue,

codex-rs/core/src/config.rs

@@ -169,14 +169,15 @@ pub struct Config {
     /// model family's default preference.
     pub include_apply_patch_tool: bool,
 
-    /// Include the `subagent.run` tool allowing the model to invoke configured subagents.
-    pub include_subagent_tool: bool,
+    pub tools_web_search_request: bool,
 
     /// The value for the `originator` header included with Responses API requests.
     pub responses_originator_header: String,
 
     /// If set to `true`, the API key will be signed with the `originator` header.
     pub preferred_auth_method: AuthMode,
+
+    pub use_experimental_streamable_shell_tool: bool,
 }
 
 impl Config {
@@ -472,6 +473,8 @@ pub struct ConfigToml {
     /// Experimental path to a file whose contents replace the built-in BASE_INSTRUCTIONS.
     pub experimental_instructions_file: Option<PathBuf>,
 
+    pub experimental_use_exec_command_tool: Option<bool>,
+
     /// The value for the `originator` header included with Responses API requests.
     pub responses_originator_header_internal_override: Option<String>,
 
@@ -480,8 +483,8 @@ pub struct ConfigToml {
     /// If set to `true`, the API key will be signed with the `originator` header.
     pub preferred_auth_method: Option<AuthMode>,
 
-    /// Include the `subagent.run` tool allowing the model to invoke configured subagents.
-    pub include_subagent_tool: Option<bool>,
+    /// Nested tools section for feature toggles
+    pub tools: Option<ToolsToml>,
 }
 
 #[derive(Deserialize, Debug, Clone, PartialEq, Eq)]
@@ -489,6 +492,13 @@ pub struct ProjectConfig {
     pub trust_level: Option<String>,
 }
 
+#[derive(Deserialize, Debug, Clone, Default)]
+pub struct ToolsToml {
+    // Renamed from `web_search_request`; keep alias for backwards compatibility.
+    #[serde(default, alias = "web_search_request")]
+    pub web_search: Option<bool>,
+}
+
 impl ConfigToml {
     /// Derive the effective sandbox policy from the configuration.
     fn derive_sandbox_policy(&self, sandbox_mode_override: Option<SandboxMode>) -> SandboxPolicy {
@@ -576,9 +586,9 @@ pub struct ConfigOverrides {
     pub base_instructions: Option<String>,
     pub include_plan_tool: Option<bool>,
     pub include_apply_patch_tool: Option<bool>,
-    pub include_subagent_tool: Option<bool>,
     pub disable_response_storage: Option<bool>,
     pub show_raw_agent_reasoning: Option<bool>,
+    pub tools_web_search_request: Option<bool>,
 }
 
 impl Config {
@@ -603,9 +613,9 @@ impl Config {
             base_instructions,
             include_plan_tool,
             include_apply_patch_tool,
-            include_subagent_tool,
             disable_response_storage,
             show_raw_agent_reasoning,
+            tools_web_search_request: override_tools_web_search_request,
         } = overrides;
 
         let config_profile = match config_profile_key.as_ref().or(cfg.profile.as_ref()) {
@@ -644,7 +654,7 @@ impl Config {
             })?
             .clone();
 
-        let shell_environment_policy = cfg.shell_environment_policy.into();
+        let shell_environment_policy = cfg.shell_environment_policy.clone().into();
 
         let resolved_cwd = {
             use std::env;
@@ -665,7 +675,11 @@ impl Config {
             }
         };
 
-        let history = cfg.history.unwrap_or_default();
+        let history = cfg.history.clone().unwrap_or_default();
+
+        let tools_web_search_request = override_tools_web_search_request
+            .or(cfg.tools.as_ref().and_then(|t| t.web_search))
+            .unwrap_or(false);
 
         let model = model
             .or(config_profile.model)
@@ -739,7 +753,7 @@ impl Config {
             codex_home,
             history,
             file_opener: cfg.file_opener.unwrap_or(UriBasedFileOpener::VsCode),
-            tui: cfg.tui.unwrap_or_default(),
+            tui: cfg.tui.clone().unwrap_or_default(),
             codex_linux_sandbox_exe,
 
             hide_agent_reasoning: cfg.hide_agent_reasoning.unwrap_or(false),
@@ -758,19 +772,18 @@ impl Config {
             model_verbosity: config_profile.model_verbosity.or(cfg.model_verbosity),
             chatgpt_base_url: config_profile
                 .chatgpt_base_url
-                .or(cfg.chatgpt_base_url)
+                .or(cfg.chatgpt_base_url.clone())
                 .unwrap_or("https://chatgpt.com/backend-api/".to_string()),
 
             experimental_resume,
             include_plan_tool: include_plan_tool.unwrap_or(false),
             include_apply_patch_tool: include_apply_patch_tool.unwrap_or(false),
-            include_subagent_tool: config_profile
-                .include_subagent_tool
-                .or(cfg.include_subagent_tool)
-                .or(include_subagent_tool)
-                .unwrap_or(false),
+            tools_web_search_request,
             responses_originator_header,
             preferred_auth_method: cfg.preferred_auth_method.unwrap_or(AuthMode::ChatGPT),
+            use_experimental_streamable_shell_tool: cfg
+                .experimental_use_exec_command_tool
+                .unwrap_or(false),
         };
         Ok(config)
     }
@@ -1135,9 +1148,10 @@ disable_response_storage = true
                 base_instructions: None,
                 include_plan_tool: false,
                 include_apply_patch_tool: false,
-                include_subagent_tool: false,
+                tools_web_search_request: false,
                 responses_originator_header: "codex_cli_rs".to_string(),
                 preferred_auth_method: AuthMode::ChatGPT,
+                use_experimental_streamable_shell_tool: false,
             },
             o3_profile_config
         );
@@ -1190,9 +1204,10 @@ disable_response_storage = true
             base_instructions: None,
             include_plan_tool: false,
             include_apply_patch_tool: false,
-            include_subagent_tool: false,
+            tools_web_search_request: false,
             responses_originator_header: "codex_cli_rs".to_string(),
             preferred_auth_method: AuthMode::ChatGPT,
+            use_experimental_streamable_shell_tool: false,
         };
 
         assert_eq!(expected_gpt3_profile_config, gpt3_profile_config);
@@ -1260,9 +1275,10 @@ disable_response_storage = true
             base_instructions: None,
             include_plan_tool: false,
             include_apply_patch_tool: false,
-            include_subagent_tool: false,
+            tools_web_search_request: false,
             responses_originator_header: "codex_cli_rs".to_string(),
             preferred_auth_method: AuthMode::ChatGPT,
+            use_experimental_streamable_shell_tool: false,
         };
 
         assert_eq!(expected_zdr_profile_config, zdr_profile_config);

codex-rs/core/src/config_profile.rs

@@ -21,6 +21,4 @@ pub struct ConfigProfile {
     pub model_verbosity: Option<Verbosity>,
     pub chatgpt_base_url: Option<String>,
     pub experimental_instructions_file: Option<PathBuf>,
-    /// Include the `subagent.run` tool allowing the model to invoke configured subagents.
-    pub include_subagent_tool: Option<bool>,
 }

codex-rs/core/src/exec_command/exec_command_params.rs

@@ -0,0 +1,57 @@
+use serde::Deserialize;
+use serde::Serialize;
+
+use crate::exec_command::session_id::SessionId;
+
+#[derive(Debug, Clone, Deserialize)]
+pub struct ExecCommandParams {
+    pub(crate) cmd: String,
+
+    #[serde(default = "default_yield_time")]
+    pub(crate) yield_time_ms: u64,
+
+    #[serde(default = "max_output_tokens")]
+    pub(crate) max_output_tokens: u64,
+
+    #[serde(default = "default_shell")]
+    pub(crate) shell: String,
+
+    #[serde(default = "default_login")]
+    pub(crate) login: bool,
+}
+
+fn default_yield_time() -> u64 {
+    10_000
+}
+
+fn max_output_tokens() -> u64 {
+    10_000
+}
+
+fn default_login() -> bool {
+    true
+}
+
+fn default_shell() -> String {
+    "/bin/bash".to_string()
+}
+
+#[derive(Debug, Deserialize, Serialize)]
+pub struct WriteStdinParams {
+    pub(crate) session_id: SessionId,
+    pub(crate) chars: String,
+
+    #[serde(default = "write_stdin_default_yield_time_ms")]
+    pub(crate) yield_time_ms: u64,
+
+    #[serde(default = "write_stdin_default_max_output_tokens")]
+    pub(crate) max_output_tokens: u64,
+}
+
+fn write_stdin_default_yield_time_ms() -> u64 {
+    250
+}
+
+fn write_stdin_default_max_output_tokens() -> u64 {
+    10_000
+}

codex-rs/core/src/exec_command/exec_command_session.rs

@@ -0,0 +1,83 @@
+use std::sync::Mutex as StdMutex;
+
+use tokio::sync::broadcast;
+use tokio::sync::mpsc;
+use tokio::task::JoinHandle;
+
+#[derive(Debug)]
+pub(crate) struct ExecCommandSession {
+    /// Queue for writing bytes to the process stdin (PTY master write side).
+    writer_tx: mpsc::Sender<Vec<u8>>,
+    /// Broadcast stream of output chunks read from the PTY. New subscribers
+    /// receive only chunks emitted after they subscribe.
+    output_tx: broadcast::Sender<Vec<u8>>,
+
+    /// Child killer handle for termination on drop (can signal independently
+    /// of a thread blocked in `.wait()`).
+    killer: StdMutex<Option<Box<dyn portable_pty::ChildKiller + Send + Sync>>>,
+
+    /// JoinHandle for the blocking PTY reader task.
+    reader_handle: StdMutex<Option<JoinHandle<()>>>,
+
+    /// JoinHandle for the stdin writer task.
+    writer_handle: StdMutex<Option<JoinHandle<()>>>,
+
+    /// JoinHandle for the child wait task.
+    wait_handle: StdMutex<Option<JoinHandle<()>>>,
+}
+
+impl ExecCommandSession {
+    pub(crate) fn new(
+        writer_tx: mpsc::Sender<Vec<u8>>,
+        output_tx: broadcast::Sender<Vec<u8>>,
+        killer: Box<dyn portable_pty::ChildKiller + Send + Sync>,
+        reader_handle: JoinHandle<()>,
+        writer_handle: JoinHandle<()>,
+        wait_handle: JoinHandle<()>,
+    ) -> Self {
+        Self {
+            writer_tx,
+            output_tx,
+            killer: StdMutex::new(Some(killer)),
+            reader_handle: StdMutex::new(Some(reader_handle)),
+            writer_handle: StdMutex::new(Some(writer_handle)),
+            wait_handle: StdMutex::new(Some(wait_handle)),
+        }
+    }
+
+    pub(crate) fn writer_sender(&self) -> mpsc::Sender<Vec<u8>> {
+        self.writer_tx.clone()
+    }
+
+    pub(crate) fn output_receiver(&self) -> broadcast::Receiver<Vec<u8>> {
+        self.output_tx.subscribe()
+    }
+}
+
+impl Drop for ExecCommandSession {
+    fn drop(&mut self) {
+        // Best-effort: terminate child first so blocking tasks can complete.
+        if let Ok(mut killer_opt) = self.killer.lock()
+            && let Some(mut killer) = killer_opt.take()
+        {
+            let _ = killer.kill();
+        }
+
+        // Abort background tasks; they may already have exited after kill.
+        if let Ok(mut h) = self.reader_handle.lock()
+            && let Some(handle) = h.take()
+        {
+            handle.abort();
+        }
+        if let Ok(mut h) = self.writer_handle.lock()
+            && let Some(handle) = h.take()
+        {
+            handle.abort();
+        }
+        if let Ok(mut h) = self.wait_handle.lock()
+            && let Some(handle) = h.take()
+        {
+            handle.abort();
+        }
+    }
+}

codex-rs/core/src/exec_command/mod.rs

@@ -0,0 +1,14 @@
+mod exec_command_params;
+mod exec_command_session;
+mod responses_api;
+mod session_id;
+mod session_manager;
+
+pub use exec_command_params::ExecCommandParams;
+pub use exec_command_params::WriteStdinParams;
+pub use responses_api::EXEC_COMMAND_TOOL_NAME;
+pub use responses_api::WRITE_STDIN_TOOL_NAME;
+pub use responses_api::create_exec_command_tool_for_responses_api;
+pub use responses_api::create_write_stdin_tool_for_responses_api;
+pub use session_manager::SessionManager as ExecSessionManager;
+pub use session_manager::result_into_payload;

codex-rs/core/src/exec_command/responses_api.rs

@@ -0,0 +1,98 @@
+use std::collections::BTreeMap;
+
+use crate::openai_tools::JsonSchema;
+use crate::openai_tools::ResponsesApiTool;
+
+pub const EXEC_COMMAND_TOOL_NAME: &str = "exec_command";
+pub const WRITE_STDIN_TOOL_NAME: &str = "write_stdin";
+
+pub fn create_exec_command_tool_for_responses_api() -> ResponsesApiTool {
+    let mut properties = BTreeMap::<String, JsonSchema>::new();
+    properties.insert(
+        "cmd".to_string(),
+        JsonSchema::String {
+            description: Some("The shell command to execute.".to_string()),
+        },
+    );
+    properties.insert(
+        "yield_time_ms".to_string(),
+        JsonSchema::Number {
+            description: Some("The maximum time in milliseconds to wait for output.".to_string()),
+        },
+    );
+    properties.insert(
+        "max_output_tokens".to_string(),
+        JsonSchema::Number {
+            description: Some("The maximum number of tokens to output.".to_string()),
+        },
+    );
+    properties.insert(
+        "shell".to_string(),
+        JsonSchema::String {
+            description: Some("The shell to use. Defaults to \"/bin/bash\".".to_string()),
+        },
+    );
+    properties.insert(
+        "login".to_string(),
+        JsonSchema::Boolean {
+            description: Some(
+                "Whether to run the command as a login shell. Defaults to true.".to_string(),
+            ),
+        },
+    );
+
+    ResponsesApiTool {
+        name: EXEC_COMMAND_TOOL_NAME.to_owned(),
+        description: r#"Execute shell commands on the local machine with streaming output."#
+            .to_string(),
+        strict: false,
+        parameters: JsonSchema::Object {
+            properties,
+            required: Some(vec!["cmd".to_string()]),
+            additional_properties: Some(false),
+        },
+    }
+}
+
+pub fn create_write_stdin_tool_for_responses_api() -> ResponsesApiTool {
+    let mut properties = BTreeMap::<String, JsonSchema>::new();
+    properties.insert(
+        "session_id".to_string(),
+        JsonSchema::Number {
+            description: Some("The ID of the exec_command session.".to_string()),
+        },
+    );
+    properties.insert(
+        "chars".to_string(),
+        JsonSchema::String {
+            description: Some("The characters to write to stdin.".to_string()),
+        },
+    );
+    properties.insert(
+        "yield_time_ms".to_string(),
+        JsonSchema::Number {
+            description: Some(
+                "The maximum time in milliseconds to wait for output after writing.".to_string(),
+            ),
+        },
+    );
+    properties.insert(
+        "max_output_tokens".to_string(),
+        JsonSchema::Number {
+            description: Some("The maximum number of tokens to output.".to_string()),
+        },
+    );
+
+    ResponsesApiTool {
+        name: WRITE_STDIN_TOOL_NAME.to_owned(),
+        description: r#"Write characters to an exec session's stdin. Returns all stdout+stderr received within yield_time_ms.
+Can write control characters (\u0003 for Ctrl-C), or an empty string to just poll stdout+stderr."#
+            .to_string(),
+        strict: false,
+        parameters: JsonSchema::Object {
+            properties,
+            required: Some(vec!["session_id".to_string(), "chars".to_string()]),
+            additional_properties: Some(false),
+        },
+    }
+}

codex-rs/core/src/exec_command/session_id.rs

@@ -0,0 +1,5 @@
+use serde::Deserialize;
+use serde::Serialize;
+
+#[derive(Debug, Clone, Copy, PartialEq, Eq, Hash, Serialize, Deserialize)]
+pub(crate) struct SessionId(pub u32);

codex-rs/core/src/exec_command/session_manager.rs

@@ -0,0 +1,674 @@
+use std::collections::HashMap;
+use std::io::ErrorKind;
+use std::io::Read;
+use std::sync::Arc;
+use std::sync::Mutex as StdMutex;
+use std::sync::atomic::AtomicU32;
+
+use portable_pty::CommandBuilder;
+use portable_pty::PtySize;
+use portable_pty::native_pty_system;
+use tokio::sync::Mutex;
+use tokio::sync::mpsc;
+use tokio::sync::oneshot;
+use tokio::time::Duration;
+use tokio::time::Instant;
+use tokio::time::timeout;
+
+use crate::exec_command::exec_command_params::ExecCommandParams;
+use crate::exec_command::exec_command_params::WriteStdinParams;
+use crate::exec_command::exec_command_session::ExecCommandSession;
+use crate::exec_command::session_id::SessionId;
+use codex_protocol::models::FunctionCallOutputPayload;
+
+#[derive(Debug, Default)]
+pub struct SessionManager {
+    next_session_id: AtomicU32,
+    sessions: Mutex<HashMap<SessionId, ExecCommandSession>>,
+}
+
+#[derive(Debug)]
+pub struct ExecCommandOutput {
+    wall_time: Duration,
+    exit_status: ExitStatus,
+    original_token_count: Option<u64>,
+    output: String,
+}
+
+impl ExecCommandOutput {
+    fn to_text_output(&self) -> String {
+        let wall_time_secs = self.wall_time.as_secs_f32();
+        let termination_status = match self.exit_status {
+            ExitStatus::Exited(code) => format!("Process exited with code {code}"),
+            ExitStatus::Ongoing(session_id) => {
+                format!("Process running with session ID {}", session_id.0)
+            }
+        };
+        let truncation_status = match self.original_token_count {
+            Some(tokens) => {
+                format!("\nWarning: truncated output (original token count: {tokens})")
+            }
+            None => "".to_string(),
+        };
+        format!(
+            r#"Wall time: {wall_time_secs:.3} seconds
+{termination_status}{truncation_status}
+Output:
+{output}"#,
+            output = self.output
+        )
+    }
+}
+
+#[derive(Debug)]
+pub enum ExitStatus {
+    Exited(i32),
+    Ongoing(SessionId),
+}
+
+pub fn result_into_payload(result: Result<ExecCommandOutput, String>) -> FunctionCallOutputPayload {
+    match result {
+        Ok(output) => FunctionCallOutputPayload {
+            content: output.to_text_output(),
+            success: Some(true),
+        },
+        Err(err) => FunctionCallOutputPayload {
+            content: err,
+            success: Some(false),
+        },
+    }
+}
+
+impl SessionManager {
+    /// Processes the request and is required to send a response via `outgoing`.
+    pub async fn handle_exec_command_request(
+        &self,
+        params: ExecCommandParams,
+    ) -> Result<ExecCommandOutput, String> {
+        // Allocate a session id.
+        let session_id = SessionId(
+            self.next_session_id
+                .fetch_add(1, std::sync::atomic::Ordering::SeqCst),
+        );
+
+        let (session, mut exit_rx) =
+            create_exec_command_session(params.clone())
+                .await
+                .map_err(|err| {
+                    format!(
+                        "failed to create exec command session for session id {}: {err}",
+                        session_id.0
+                    )
+                })?;
+
+        // Insert into session map.
+        let mut output_rx = session.output_receiver();
+        self.sessions.lock().await.insert(session_id, session);
+
+        // Collect output until either timeout expires or process exits.
+        // Do not cap during collection; truncate at the end if needed.
+        // Use a modest initial capacity to avoid large preallocation.
+        let cap_bytes_u64 = params.max_output_tokens.saturating_mul(4);
+        let cap_bytes: usize = cap_bytes_u64.min(usize::MAX as u64) as usize;
+        let mut collected: Vec<u8> = Vec::with_capacity(4096);
+
+        let start_time = Instant::now();
+        let deadline = start_time + Duration::from_millis(params.yield_time_ms);
+        let mut exit_code: Option<i32> = None;
+
+        loop {
+            if Instant::now() >= deadline {
+                break;
+            }
+            let remaining = deadline.saturating_duration_since(Instant::now());
+            tokio::select! {
+                biased;
+                exit = &mut exit_rx => {
+                    exit_code = exit.ok();
+                    // Small grace period to pull remaining buffered output
+                    let grace_deadline = Instant::now() + Duration::from_millis(25);
+                    while Instant::now() < grace_deadline {
+                        match timeout(Duration::from_millis(1), output_rx.recv()).await {
+                            Ok(Ok(chunk)) => {
+                                collected.extend_from_slice(&chunk);
+                            }
+                            Ok(Err(tokio::sync::broadcast::error::RecvError::Lagged(_))) => {
+                                // Skip missed messages; keep trying within grace period.
+                                continue;
+                            }
+                            Ok(Err(tokio::sync::broadcast::error::RecvError::Closed)) => break,
+                            Err(_) => break,
+                        }
+                    }
+                    break;
+                }
+                chunk = timeout(remaining, output_rx.recv()) => {
+                    match chunk {
+                        Ok(Ok(chunk)) => {
+                            collected.extend_from_slice(&chunk);
+                        }
+                        Ok(Err(tokio::sync::broadcast::error::RecvError::Lagged(_))) => {
+                            // Skip missed messages; continue collecting fresh output.
+                        }
+                        Ok(Err(tokio::sync::broadcast::error::RecvError::Closed)) => { break; }
+                        Err(_) => { break; }
+                    }
+                }
+            }
+        }
+
+        let output = String::from_utf8_lossy(&collected).to_string();
+
+        let exit_status = if let Some(code) = exit_code {
+            ExitStatus::Exited(code)
+        } else {
+            ExitStatus::Ongoing(session_id)
+        };
+
+        // If output exceeds cap, truncate the middle and record original token estimate.
+        let (output, original_token_count) = truncate_middle(&output, cap_bytes);
+        Ok(ExecCommandOutput {
+            wall_time: Instant::now().duration_since(start_time),
+            exit_status,
+            original_token_count,
+            output,
+        })
+    }
+
+    /// Write characters to a session's stdin and collect combined output for up to `yield_time_ms`.
+    pub async fn handle_write_stdin_request(
+        &self,
+        params: WriteStdinParams,
+    ) -> Result<ExecCommandOutput, String> {
+        let WriteStdinParams {
+            session_id,
+            chars,
+            yield_time_ms,
+            max_output_tokens,
+        } = params;
+
+        // Grab handles without holding the sessions lock across await points.
+        let (writer_tx, mut output_rx) = {
+            let sessions = self.sessions.lock().await;
+            match sessions.get(&session_id) {
+                Some(session) => (session.writer_sender(), session.output_receiver()),
+                None => {
+                    return Err(format!("unknown session id {}", session_id.0));
+                }
+            }
+        };
+
+        // Write stdin if provided.
+        if !chars.is_empty() && writer_tx.send(chars.into_bytes()).await.is_err() {
+            return Err("failed to write to stdin".to_string());
+        }
+
+        // Collect output up to yield_time_ms, truncating to max_output_tokens bytes.
+        let mut collected: Vec<u8> = Vec::with_capacity(4096);
+        let start_time = Instant::now();
+        let deadline = start_time + Duration::from_millis(yield_time_ms);
+        loop {
+            let now = Instant::now();
+            if now >= deadline {
+                break;
+            }
+            let remaining = deadline - now;
+            match timeout(remaining, output_rx.recv()).await {
+                Ok(Ok(chunk)) => {
+                    // Collect all output within the time budget; truncate at the end.
+                    collected.extend_from_slice(&chunk);
+                }
+                Ok(Err(tokio::sync::broadcast::error::RecvError::Lagged(_))) => {
+                    // Skip missed messages; continue collecting fresh output.
+                }
+                Ok(Err(tokio::sync::broadcast::error::RecvError::Closed)) => break,
+                Err(_) => break, // timeout
+            }
+        }
+
+        // Return structured output, truncating middle if over cap.
+        let output = String::from_utf8_lossy(&collected).to_string();
+        let cap_bytes_u64 = max_output_tokens.saturating_mul(4);
+        let cap_bytes: usize = cap_bytes_u64.min(usize::MAX as u64) as usize;
+        let (output, original_token_count) = truncate_middle(&output, cap_bytes);
+        Ok(ExecCommandOutput {
+            wall_time: Instant::now().duration_since(start_time),
+            exit_status: ExitStatus::Ongoing(session_id),
+            original_token_count,
+            output,
+        })
+    }
+}
+
+/// Spawn PTY and child process per spawn_exec_command_session logic.
+async fn create_exec_command_session(
+    params: ExecCommandParams,
+) -> anyhow::Result<(ExecCommandSession, oneshot::Receiver<i32>)> {
+    let ExecCommandParams {
+        cmd,
+        yield_time_ms: _,
+        max_output_tokens: _,
+        shell,
+        login,
+    } = params;
+
+    // Use the native pty implementation for the system
+    let pty_system = native_pty_system();
+
+    // Create a new pty
+    let pair = pty_system.openpty(PtySize {
+        rows: 24,
+        cols: 80,
+        pixel_width: 0,
+        pixel_height: 0,
+    })?;
+
+    // Spawn a shell into the pty
+    let mut command_builder = CommandBuilder::new(shell);
+    let shell_mode_opt = if login { "-lc" } else { "-c" };
+    command_builder.arg(shell_mode_opt);
+    command_builder.arg(cmd);
+
+    let mut child = pair.slave.spawn_command(command_builder)?;
+    // Obtain a killer that can signal the process independently of `.wait()`.
+    let killer = child.clone_killer();
+
+    // Channel to forward write requests to the PTY writer.
+    let (writer_tx, mut writer_rx) = mpsc::channel::<Vec<u8>>(128);
+    // Broadcast for streaming PTY output to readers: subscribers receive from subscription time.
+    let (output_tx, _) = tokio::sync::broadcast::channel::<Vec<u8>>(256);
+
+    // Reader task: drain PTY and forward chunks to output channel.
+    let mut reader = pair.master.try_clone_reader()?;
+    let output_tx_clone = output_tx.clone();
+    let reader_handle = tokio::task::spawn_blocking(move || {
+        let mut buf = [0u8; 8192];
+        loop {
+            match reader.read(&mut buf) {
+                Ok(0) => break, // EOF
+                Ok(n) => {
+                    // Forward to broadcast; best-effort if there are subscribers.
+                    let _ = output_tx_clone.send(buf[..n].to_vec());
+                }
+                Err(ref e) if e.kind() == ErrorKind::Interrupted => {
+                    // Retry on EINTR
+                    continue;
+                }
+                Err(ref e) if e.kind() == ErrorKind::WouldBlock => {
+                    // We're in a blocking thread; back off briefly and retry.
+                    std::thread::sleep(Duration::from_millis(5));
+                    continue;
+                }
+                Err(_) => break,
+            }
+        }
+    });
+
+    // Writer task: apply stdin writes to the PTY writer.
+    let writer = pair.master.take_writer()?;
+    let writer = Arc::new(StdMutex::new(writer));
+    let writer_handle = tokio::spawn({
+        let writer = writer.clone();
+        async move {
+            while let Some(bytes) = writer_rx.recv().await {
+                let writer = writer.clone();
+                // Perform blocking write on a blocking thread.
+                let _ = tokio::task::spawn_blocking(move || {
+                    if let Ok(mut guard) = writer.lock() {
+                        use std::io::Write;
+                        let _ = guard.write_all(&bytes);
+                        let _ = guard.flush();
+                    }
+                })
+                .await;
+            }
+        }
+    });
+
+    // Keep the child alive until it exits, then signal exit code.
+    let (exit_tx, exit_rx) = oneshot::channel::<i32>();
+    let wait_handle = tokio::task::spawn_blocking(move || {
+        let code = match child.wait() {
+            Ok(status) => status.exit_code() as i32,
+            Err(_) => -1,
+        };
+        let _ = exit_tx.send(code);
+    });
+
+    // Create and store the session with channels.
+    let session = ExecCommandSession::new(
+        writer_tx,
+        output_tx,
+        killer,
+        reader_handle,
+        writer_handle,
+        wait_handle,
+    );
+    Ok((session, exit_rx))
+}
+
+/// Truncate the middle of a UTF-8 string to at most `max_bytes` bytes,
+/// preserving the beginning and the end. Returns the possibly truncated
+/// string and `Some(original_token_count)` (estimated at 4 bytes/token)
+/// if truncation occurred; otherwise returns the original string and `None`.
+fn truncate_middle(s: &str, max_bytes: usize) -> (String, Option<u64>) {
+    // No truncation needed
+    if s.len() <= max_bytes {
+        return (s.to_string(), None);
+    }
+    let est_tokens = (s.len() as u64).div_ceil(4);
+    if max_bytes == 0 {
+        // Cannot keep any content; still return a full marker (never truncated).
+        return (
+            format!("…{} tokens truncated…", est_tokens),
+            Some(est_tokens),
+        );
+    }
+
+    // Helper to truncate a string to a given byte length on a char boundary.
+    fn truncate_on_boundary(input: &str, max_len: usize) -> &str {
+        if input.len() <= max_len {
+            return input;
+        }
+        let mut end = max_len;
+        while end > 0 && !input.is_char_boundary(end) {
+            end -= 1;
+        }
+        &input[..end]
+    }
+
+    // Given a left/right budget, prefer newline boundaries; otherwise fall back
+    // to UTF-8 char boundaries.
+    fn pick_prefix_end(s: &str, left_budget: usize) -> usize {
+        if let Some(head) = s.get(..left_budget)
+            && let Some(i) = head.rfind('\n')
+        {
+            return i + 1; // keep the newline so suffix starts on a fresh line
+        }
+        truncate_on_boundary(s, left_budget).len()
+    }
+
+    fn pick_suffix_start(s: &str, right_budget: usize) -> usize {
+        let start_tail = s.len().saturating_sub(right_budget);
+        if let Some(tail) = s.get(start_tail..)
+            && let Some(i) = tail.find('\n')
+        {
+            return start_tail + i + 1; // start after newline
+        }
+        // Fall back to a char boundary at or after start_tail.
+        let mut idx = start_tail.min(s.len());
+        while idx < s.len() && !s.is_char_boundary(idx) {
+            idx += 1;
+        }
+        idx
+    }
+
+    // Refine marker length and budgets until stable. Marker is never truncated.
+    let mut guess_tokens = est_tokens; // worst-case: everything truncated
+    for _ in 0..4 {
+        let marker = format!("…{} tokens truncated…", guess_tokens);
+        let marker_len = marker.len();
+        let keep_budget = max_bytes.saturating_sub(marker_len);
+        if keep_budget == 0 {
+            // No room for any content within the cap; return a full, untruncated marker
+            // that reflects the entire truncated content.
+            return (
+                format!("…{} tokens truncated…", est_tokens),
+                Some(est_tokens),
+            );
+        }
+
+        let left_budget = keep_budget / 2;
+        let right_budget = keep_budget - left_budget;
+        let prefix_end = pick_prefix_end(s, left_budget);
+        let mut suffix_start = pick_suffix_start(s, right_budget);
+        if suffix_start < prefix_end {
+            suffix_start = prefix_end;
+        }
+        let kept_content_bytes = prefix_end + (s.len() - suffix_start);
+        let truncated_content_bytes = s.len().saturating_sub(kept_content_bytes);
+        let new_tokens = (truncated_content_bytes as u64).div_ceil(4);
+        if new_tokens == guess_tokens {
+            let mut out = String::with_capacity(marker_len + kept_content_bytes + 1);
+            out.push_str(&s[..prefix_end]);
+            out.push_str(&marker);
+            // Place marker on its own line for symmetry when we keep line boundaries.
+            out.push('\n');
+            out.push_str(&s[suffix_start..]);
+            return (out, Some(est_tokens));
+        }
+        guess_tokens = new_tokens;
+    }
+
+    // Fallback: use last guess to build output.
+    let marker = format!("…{} tokens truncated…", guess_tokens);
+    let marker_len = marker.len();
+    let keep_budget = max_bytes.saturating_sub(marker_len);
+    if keep_budget == 0 {
+        return (
+            format!("…{} tokens truncated…", est_tokens),
+            Some(est_tokens),
+        );
+    }
+    let left_budget = keep_budget / 2;
+    let right_budget = keep_budget - left_budget;
+    let prefix_end = pick_prefix_end(s, left_budget);
+    let suffix_start = pick_suffix_start(s, right_budget);
+    let mut out = String::with_capacity(marker_len + prefix_end + (s.len() - suffix_start) + 1);
+    out.push_str(&s[..prefix_end]);
+    out.push_str(&marker);
+    out.push('\n');
+    out.push_str(&s[suffix_start..]);
+    (out, Some(est_tokens))
+}
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+    use crate::exec_command::session_id::SessionId;
+
+    /// Test that verifies that [`SessionManager::handle_exec_command_request()`]
+    /// and [`SessionManager::handle_write_stdin_request()`] work as expected
+    /// in the presence of a process that never terminates (but produces
+    /// output continuously).
+    #[cfg(unix)]
+    #[allow(clippy::print_stderr)]
+    #[tokio::test(flavor = "multi_thread", worker_threads = 4)]
+    async fn session_manager_streams_and_truncates_from_now() {
+        use crate::exec_command::exec_command_params::ExecCommandParams;
+        use crate::exec_command::exec_command_params::WriteStdinParams;
+        use tokio::time::sleep;
+
+        let session_manager = SessionManager::default();
+        // Long-running loop that prints an increasing counter every ~100ms.
+        // Use Python for a portable, reliable sleep across shells/PTYs.
+        let cmd = r#"python3 - <<'PY'
+import sys, time
+count = 0
+while True:
+    print(count)
+    sys.stdout.flush()
+    count += 100
+    time.sleep(0.1)
+PY"#
+        .to_string();
+
+        // Start the session and collect ~3s of output.
+        let params = ExecCommandParams {
+            cmd,
+            yield_time_ms: 3_000,
+            max_output_tokens: 1_000, // large enough to avoid truncation here
+            shell: "/bin/bash".to_string(),
+            login: false,
+        };
+        let initial_output = match session_manager
+            .handle_exec_command_request(params.clone())
+            .await
+        {
+            Ok(v) => v,
+            Err(e) => {
+                // PTY may be restricted in some sandboxes; skip in that case.
+                if e.contains("openpty") || e.contains("Operation not permitted") {
+                    eprintln!("skipping test due to restricted PTY: {e}");
+                    return;
+                }
+                panic!("exec request failed unexpectedly: {e}");
+            }
+        };
+        eprintln!("initial output: {initial_output:?}");
+
+        // Should be ongoing (we launched a never-ending loop).
+        let session_id = match initial_output.exit_status {
+            ExitStatus::Ongoing(id) => id,
+            _ => panic!("expected ongoing session"),
+        };
+
+        // Parse the numeric lines and get the max observed value in the first window.
+        let first_nums = extract_monotonic_numbers(&initial_output.output);
+        assert!(
+            !first_nums.is_empty(),
+            "expected some output from first window"
+        );
+        let first_max = *first_nums.iter().max().unwrap();
+
+        // Wait ~4s so counters progress while we're not reading.
+        sleep(Duration::from_millis(4_000)).await;
+
+        // Now read ~3s of output "from now" only.
+        // Use a small token cap so truncation occurs and we test middle truncation.
+        let write_params = WriteStdinParams {
+            session_id,
+            chars: String::new(),
+            yield_time_ms: 3_000,
+            max_output_tokens: 16, // 16 tokens ~= 64 bytes -> likely truncation
+        };
+        let second = session_manager
+            .handle_write_stdin_request(write_params)
+            .await
+            .expect("write stdin should succeed");
+
+        // Verify truncation metadata and size bound (cap is tokens*4 bytes).
+        assert!(second.original_token_count.is_some());
+        let cap_bytes = (16u64 * 4) as usize;
+        assert!(second.output.len() <= cap_bytes);
+        // New middle marker should be present.
+        assert!(
+            second.output.contains("tokens truncated") && second.output.contains('…'),
+            "expected truncation marker in output, got: {}",
+            second.output
+        );
+
+        // Minimal freshness check: the earliest number we see in the second window
+        // should be significantly larger than the last from the first window.
+        let second_nums = extract_monotonic_numbers(&second.output);
+        assert!(
+            !second_nums.is_empty(),
+            "expected some numeric output from second window"
+        );
+        let second_min = *second_nums.iter().min().unwrap();
+
+        // We slept 4 seconds (~40 ticks at 100ms/tick, each +100), so expect
+        // an increase of roughly 4000 or more. Allow a generous margin.
+        assert!(
+            second_min >= first_max + 2000,
+            "second_min={second_min} first_max={first_max}",
+        );
+    }
+
+    #[cfg(unix)]
+    fn extract_monotonic_numbers(s: &str) -> Vec<i64> {
+        s.lines()
+            .filter_map(|line| {
+                if !line.is_empty()
+                    && line.chars().all(|c| c.is_ascii_digit())
+                    && let Ok(n) = line.parse::<i64>()
+                {
+                    // Our generator increments by 100; ignore spurious fragments.
+                    if n % 100 == 0 {
+                        return Some(n);
+                    }
+                }
+                None
+            })
+            .collect()
+    }
+
+    #[test]
+    fn to_text_output_exited_no_truncation() {
+        let out = ExecCommandOutput {
+            wall_time: Duration::from_millis(1234),
+            exit_status: ExitStatus::Exited(0),
+            original_token_count: None,
+            output: "hello".to_string(),
+        };
+        let text = out.to_text_output();
+        let expected = r#"Wall time: 1.234 seconds
+Process exited with code 0
+Output:
+hello"#;
+        assert_eq!(expected, text);
+    }
+
+    #[test]
+    fn to_text_output_ongoing_with_truncation() {
+        let out = ExecCommandOutput {
+            wall_time: Duration::from_millis(500),
+            exit_status: ExitStatus::Ongoing(SessionId(42)),
+            original_token_count: Some(1000),
+            output: "abc".to_string(),
+        };
+        let text = out.to_text_output();
+        let expected = r#"Wall time: 0.500 seconds
+Process running with session ID 42
+Warning: truncated output (original token count: 1000)
+Output:
+abc"#;
+        assert_eq!(expected, text);
+    }
+
+    #[test]
+    fn truncate_middle_no_newlines_fallback() {
+        // A long string with no newlines that exceeds the cap.
+        let s = "abcdefghijklmnopqrstuvwxyz0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZ";
+        let max_bytes = 16; // force truncation
+        let (out, original) = truncate_middle(s, max_bytes);
+        // For very small caps, we return the full, untruncated marker,
+        // even if it exceeds the cap.
+        assert_eq!(out, "…16 tokens truncated…");
+        // Original string length is 62 bytes => ceil(62/4) = 16 tokens.
+        assert_eq!(original, Some(16));
+    }
+
+    #[test]
+    fn truncate_middle_prefers_newline_boundaries() {
+        // Build a multi-line string of 20 numbered lines (each "NNN\n").
+        let mut s = String::new();
+        for i in 1..=20 {
+            s.push_str(&format!("{i:03}\n"));
+        }
+        // Total length: 20 lines * 4 bytes per line = 80 bytes.
+        assert_eq!(s.len(), 80);
+
+        // Choose a cap that forces truncation while leaving room for
+        // a few lines on each side after accounting for the marker.
+        let max_bytes = 64;
+        // Expect exact output: first 4 lines, marker, last 4 lines, and correct token estimate (80/4 = 20).
+        assert_eq!(
+            truncate_middle(&s, max_bytes),
+            (
+                r#"001
+002
+003
+004
+…12 tokens truncated…
+017
+018
+019
+020
+"#
+                .to_string(),
+                Some(20)
+            )
+        );
+    }
+}

codex-rs/core/src/lib.rs

@@ -20,6 +20,7 @@ mod conversation_history;
 mod environment_context;
 pub mod error;
 pub mod exec;
+mod exec_command;
 pub mod exec_env;
 mod flags;
 pub mod git_info;
@@ -62,4 +63,3 @@ pub use codex_protocol::protocol;
 // Re-export protocol config enums to ensure call sites can use the same types
 // as those in the protocol crate when constructing protocol messages.
 pub use codex_protocol::config_types as protocol_config_types;
-pub mod subagents;

codex-rs/core/src/model_family.rs

@@ -90,7 +90,6 @@ pub fn find_family_for_model(slug: &str) -> Option<ModelFamily> {
         model_family!(
             slug, slug,
             supports_reasoning_summaries: true,
-            apply_patch_tool_type: Some(ApplyPatchToolType::Freeform),
         )
     } else if slug.starts_with("gpt-4.1") {
         model_family!(
@@ -107,7 +106,6 @@ pub fn find_family_for_model(slug: &str) -> Option<ModelFamily> {
         model_family!(
             slug, "gpt-5",
             supports_reasoning_summaries: true,
-            apply_patch_tool_type: Some(ApplyPatchToolType::Freeform),
         )
     } else {
         None

codex-rs/core/src/openai_model_info.rs

@@ -79,13 +79,13 @@ pub(crate) fn get_model_info(model_family: &ModelFamily) -> Option<ModelInfo> {
         }),
 
         "gpt-5" => Some(ModelInfo {
-            context_window: 200_000,
-            max_output_tokens: 100_000,
+            context_window: 400_000,
+            max_output_tokens: 128_000,
         }),
 
         _ if slug.starts_with("codex-") => Some(ModelInfo {
-            context_window: 200_000,
-            max_output_tokens: 100_000,
+            context_window: 400_000,
+            max_output_tokens: 128_000,
         }),
 
         _ => None,

codex-rs/core/src/openai_tools.rs

@@ -47,6 +47,8 @@ pub(crate) enum OpenAiTool {
     Function(ResponsesApiTool),
     #[serde(rename = "local_shell")]
     LocalShell {},
+    #[serde(rename = "web_search")]
+    WebSearch {},
     #[serde(rename = "custom")]
     Freeform(FreeformTool),
 }
@@ -56,31 +58,46 @@ pub enum ConfigShellToolType {
     DefaultShell,
     ShellWithRequest { sandbox_policy: SandboxPolicy },
     LocalShell,
+    StreamableShell,
 }
 
 #[derive(Debug, Clone)]
-pub struct ToolsConfig {
+pub(crate) struct ToolsConfig {
     pub shell_type: ConfigShellToolType,
     pub plan_tool: bool,
     pub apply_patch_tool_type: Option<ApplyPatchToolType>,
-    pub subagent_tool: bool,
+    pub web_search_request: bool,
+}
+
+pub(crate) struct ToolsConfigParams<'a> {
+    pub(crate) model_family: &'a ModelFamily,
+    pub(crate) approval_policy: AskForApproval,
+    pub(crate) sandbox_policy: SandboxPolicy,
+    pub(crate) include_plan_tool: bool,
+    pub(crate) include_apply_patch_tool: bool,
+    pub(crate) include_web_search_request: bool,
+    pub(crate) use_streamable_shell_tool: bool,
 }
 
 impl ToolsConfig {
-    pub fn new(
-        model_family: &ModelFamily,
-        approval_policy: AskForApproval,
-        sandbox_policy: SandboxPolicy,
-        include_plan_tool: bool,
-        include_apply_patch_tool: bool,
-        include_subagent_tool: bool,
-    ) -> Self {
-        let mut shell_type = if model_family.uses_local_shell_tool {
+    pub fn new(params: &ToolsConfigParams) -> Self {
+        let ToolsConfigParams {
+            model_family,
+            approval_policy,
+            sandbox_policy,
+            include_plan_tool,
+            include_apply_patch_tool,
+            include_web_search_request,
+            use_streamable_shell_tool,
+        } = params;
+        let mut shell_type = if *use_streamable_shell_tool {
+            ConfigShellToolType::StreamableShell
+        } else if model_family.uses_local_shell_tool {
             ConfigShellToolType::LocalShell
         } else {
             ConfigShellToolType::DefaultShell
         };
-        if matches!(approval_policy, AskForApproval::OnRequest) {
+        if matches!(approval_policy, AskForApproval::OnRequest) && !use_streamable_shell_tool {
             shell_type = ConfigShellToolType::ShellWithRequest {
                 sandbox_policy: sandbox_policy.clone(),
             }
@@ -90,7 +107,7 @@ impl ToolsConfig {
             Some(ApplyPatchToolType::Freeform) => Some(ApplyPatchToolType::Freeform),
             Some(ApplyPatchToolType::Function) => Some(ApplyPatchToolType::Function),
             None => {
-                if include_apply_patch_tool {
+                if *include_apply_patch_tool {
                     Some(ApplyPatchToolType::Freeform)
                 } else {
                     None
@@ -100,9 +117,9 @@ impl ToolsConfig {
 
         Self {
             shell_type,
-            plan_tool: include_plan_tool,
+            plan_tool: *include_plan_tool,
             apply_patch_tool_type,
-            subagent_tool: include_subagent_tool,
+            web_search_request: *include_web_search_request,
         }
     }
 }
@@ -495,6 +512,14 @@ pub(crate) fn get_openai_tools(
         ConfigShellToolType::LocalShell => {
             tools.push(OpenAiTool::LocalShell {});
         }
+        ConfigShellToolType::StreamableShell => {
+            tools.push(OpenAiTool::Function(
+                crate::exec_command::create_exec_command_tool_for_responses_api(),
+            ));
+            tools.push(OpenAiTool::Function(
+                crate::exec_command::create_write_stdin_tool_for_responses_api(),
+            ));
+        }
     }
 
     if config.plan_tool {
@@ -512,14 +537,17 @@ pub(crate) fn get_openai_tools(
         }
     }
 
-    if config.subagent_tool {
-        tracing::trace!("Adding subagent tool");
-        tools.push(crate::subagents::SUBAGENT_TOOL.clone());
-        tools.push(crate::subagents::SUBAGENT_LIST_TOOL.clone());
+    if config.web_search_request {
+        tools.push(OpenAiTool::WebSearch {});
     }
 
     if let Some(mcp_tools) = mcp_tools {
-        for (name, tool) in mcp_tools {
+        // Ensure deterministic ordering to maximize prompt cache hits.
+        // HashMap iteration order is non-deterministic, so sort by fully-qualified tool name.
+        let mut entries: Vec<(String, mcp_types::Tool)> = mcp_tools.into_iter().collect();
+        entries.sort_by(|a, b| a.0.cmp(&b.0));
+
+        for (name, tool) in entries.into_iter() {
             match mcp_tool_to_openai_tool(name.clone(), tool.clone()) {
                 Ok(converted_tool) => tools.push(OpenAiTool::Function(converted_tool)),
                 Err(e) => {
@@ -529,7 +557,6 @@ pub(crate) fn get_openai_tools(
         }
     }
 
-    tracing::trace!("Tools: {tools:?}");
     tools
 }
 
@@ -547,6 +574,7 @@ mod tests {
             .map(|tool| match tool {
                 OpenAiTool::Function(ResponsesApiTool { name, .. }) => name,
                 OpenAiTool::LocalShell {} => "local_shell",
+                OpenAiTool::WebSearch {} => "web_search",
                 OpenAiTool::Freeform(FreeformTool { name, .. }) => name,
             })
             .collect::<Vec<_>>();
@@ -568,46 +596,49 @@ mod tests {
     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,
-            false,
-            false,
-        );
+        let config = ToolsConfig::new(&ToolsConfigParams {
+            model_family: &model_family,
+            approval_policy: AskForApproval::Never,
+            sandbox_policy: SandboxPolicy::ReadOnly,
+            include_plan_tool: true,
+            include_apply_patch_tool: false,
+            include_web_search_request: true,
+            use_streamable_shell_tool: false,
+        });
         let tools = get_openai_tools(&config, Some(HashMap::new()));
 
-        assert_eq_tool_names(&tools, &["local_shell", "update_plan"]);
+        assert_eq_tool_names(&tools, &["local_shell", "update_plan", "web_search"]);
     }
 
     #[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,
-            false,
-            false,
-        );
+        let config = ToolsConfig::new(&ToolsConfigParams {
+            model_family: &model_family,
+            approval_policy: AskForApproval::Never,
+            sandbox_policy: SandboxPolicy::ReadOnly,
+            include_plan_tool: true,
+            include_apply_patch_tool: false,
+            include_web_search_request: true,
+            use_streamable_shell_tool: false,
+        });
         let tools = get_openai_tools(&config, Some(HashMap::new()));
 
-        assert_eq_tool_names(&tools, &["shell", "update_plan"]);
+        assert_eq_tool_names(&tools, &["shell", "update_plan", "web_search"]);
     }
 
     #[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,
-            false,
-            false,
-        );
+        let config = ToolsConfig::new(&ToolsConfigParams {
+            model_family: &model_family,
+            approval_policy: AskForApproval::Never,
+            sandbox_policy: SandboxPolicy::ReadOnly,
+            include_plan_tool: false,
+            include_apply_patch_tool: false,
+            include_web_search_request: true,
+            use_streamable_shell_tool: false,
+        });
         let tools = get_openai_tools(
             &config,
             Some(HashMap::from([(
@@ -629,8 +660,8 @@ mod tests {
                                     "number_property": { "type": "number" },
                                 },
                                 "required": [
-                                    "string_property",
-                                    "number_property"
+                                    "string_property".to_string(),
+                                    "number_property".to_string()
                                 ],
                                 "additionalProperties": Some(false),
                             },
@@ -646,10 +677,13 @@ mod tests {
             )])),
         );
 
-        assert_eq_tool_names(&tools, &["shell", "test_server/do_something_cool"]);
+        assert_eq_tool_names(
+            &tools,
+            &["shell", "web_search", "test_server/do_something_cool"],
+        );
 
         assert_eq!(
-            tools[1],
+            tools[2],
             OpenAiTool::Function(ResponsesApiTool {
                 name: "test_server/do_something_cool".to_string(),
                 parameters: JsonSchema::Object {
@@ -692,17 +726,93 @@ mod tests {
         );
     }
 
+    #[test]
+    fn test_get_openai_tools_mcp_tools_sorted_by_name() {
+        let model_family = find_family_for_model("o3").expect("o3 should be a valid model family");
+        let config = ToolsConfig::new(&ToolsConfigParams {
+            model_family: &model_family,
+            approval_policy: AskForApproval::Never,
+            sandbox_policy: SandboxPolicy::ReadOnly,
+            include_plan_tool: false,
+            include_apply_patch_tool: false,
+            include_web_search_request: false,
+            use_streamable_shell_tool: false,
+        });
+
+        // Intentionally construct a map with keys that would sort alphabetically.
+        let tools_map: HashMap<String, mcp_types::Tool> = HashMap::from([
+            (
+                "test_server/do".to_string(),
+                mcp_types::Tool {
+                    name: "a".to_string(),
+                    input_schema: ToolInputSchema {
+                        properties: Some(serde_json::json!({})),
+                        required: None,
+                        r#type: "object".to_string(),
+                    },
+                    output_schema: None,
+                    title: None,
+                    annotations: None,
+                    description: Some("a".to_string()),
+                },
+            ),
+            (
+                "test_server/something".to_string(),
+                mcp_types::Tool {
+                    name: "b".to_string(),
+                    input_schema: ToolInputSchema {
+                        properties: Some(serde_json::json!({})),
+                        required: None,
+                        r#type: "object".to_string(),
+                    },
+                    output_schema: None,
+                    title: None,
+                    annotations: None,
+                    description: Some("b".to_string()),
+                },
+            ),
+            (
+                "test_server/cool".to_string(),
+                mcp_types::Tool {
+                    name: "c".to_string(),
+                    input_schema: ToolInputSchema {
+                        properties: Some(serde_json::json!({})),
+                        required: None,
+                        r#type: "object".to_string(),
+                    },
+                    output_schema: None,
+                    title: None,
+                    annotations: None,
+                    description: Some("c".to_string()),
+                },
+            ),
+        ]);
+
+        let tools = get_openai_tools(&config, Some(tools_map));
+        // Expect shell first, followed by MCP tools sorted by fully-qualified name.
+        assert_eq_tool_names(
+            &tools,
+            &[
+                "shell",
+                "test_server/cool",
+                "test_server/do",
+                "test_server/something",
+            ],
+        );
+    }
+
     #[test]
     fn test_mcp_tool_property_missing_type_defaults_to_string() {
         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,
-            false,
-            false,
-        );
+        let config = ToolsConfig::new(&ToolsConfigParams {
+            model_family: &model_family,
+            approval_policy: AskForApproval::Never,
+            sandbox_policy: SandboxPolicy::ReadOnly,
+            include_plan_tool: false,
+            include_apply_patch_tool: false,
+            include_web_search_request: true,
+            use_streamable_shell_tool: false,
+        });
 
         let tools = get_openai_tools(
             &config,
@@ -727,10 +837,10 @@ mod tests {
             )])),
         );
 
-        assert_eq_tool_names(&tools, &["shell", "dash/search"]);
+        assert_eq_tool_names(&tools, &["shell", "web_search", "dash/search"]);
 
         assert_eq!(
-            tools[1],
+            tools[2],
             OpenAiTool::Function(ResponsesApiTool {
                 name: "dash/search".to_string(),
                 parameters: JsonSchema::Object {
@@ -752,14 +862,15 @@ mod tests {
     #[test]
     fn test_mcp_tool_integer_normalized_to_number() {
         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,
-            false,
-            false,
-        );
+        let config = ToolsConfig::new(&ToolsConfigParams {
+            model_family: &model_family,
+            approval_policy: AskForApproval::Never,
+            sandbox_policy: SandboxPolicy::ReadOnly,
+            include_plan_tool: false,
+            include_apply_patch_tool: false,
+            include_web_search_request: true,
+            use_streamable_shell_tool: false,
+        });
 
         let tools = get_openai_tools(
             &config,
@@ -782,9 +893,9 @@ mod tests {
             )])),
         );
 
-        assert_eq_tool_names(&tools, &["shell", "dash/paginate"]);
+        assert_eq_tool_names(&tools, &["shell", "web_search", "dash/paginate"]);
         assert_eq!(
-            tools[1],
+            tools[2],
             OpenAiTool::Function(ResponsesApiTool {
                 name: "dash/paginate".to_string(),
                 parameters: JsonSchema::Object {
@@ -804,14 +915,15 @@ mod tests {
     #[test]
     fn test_mcp_tool_array_without_items_gets_default_string_items() {
         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,
-            false,
-            false,
-        );
+        let config = ToolsConfig::new(&ToolsConfigParams {
+            model_family: &model_family,
+            approval_policy: AskForApproval::Never,
+            sandbox_policy: SandboxPolicy::ReadOnly,
+            include_plan_tool: false,
+            include_apply_patch_tool: false,
+            include_web_search_request: true,
+            use_streamable_shell_tool: false,
+        });
 
         let tools = get_openai_tools(
             &config,
@@ -834,9 +946,9 @@ mod tests {
             )])),
         );
 
-        assert_eq_tool_names(&tools, &["shell", "dash/tags"]);
+        assert_eq_tool_names(&tools, &["shell", "web_search", "dash/tags"]);
         assert_eq!(
-            tools[1],
+            tools[2],
             OpenAiTool::Function(ResponsesApiTool {
                 name: "dash/tags".to_string(),
                 parameters: JsonSchema::Object {
@@ -859,14 +971,15 @@ mod tests {
     #[test]
     fn test_mcp_tool_anyof_defaults_to_string() {
         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,
-            false,
-            false,
-        );
+        let config = ToolsConfig::new(&ToolsConfigParams {
+            model_family: &model_family,
+            approval_policy: AskForApproval::Never,
+            sandbox_policy: SandboxPolicy::ReadOnly,
+            include_plan_tool: false,
+            include_apply_patch_tool: false,
+            include_web_search_request: true,
+            use_streamable_shell_tool: false,
+        });
 
         let tools = get_openai_tools(
             &config,
@@ -889,9 +1002,9 @@ mod tests {
             )])),
         );
 
-        assert_eq_tool_names(&tools, &["shell", "dash/value"]);
+        assert_eq_tool_names(&tools, &["shell", "web_search", "dash/value"]);
         assert_eq!(
-            tools[1],
+            tools[2],
             OpenAiTool::Function(ResponsesApiTool {
                 name: "dash/value".to_string(),
                 parameters: JsonSchema::Object {

codex-rs/core/src/subagents/definition.rs

@@ -1,32 +0,0 @@
-use serde::Deserialize;
-use std::fs;
-use std::path::Path;
-
-#[derive(Debug, Clone, Deserialize)]
-pub struct SubagentDefinition {
-    pub name: String,
-    pub description: String,
-    /// Base instructions for this subagent.
-    pub instructions: String,
-    /// When not set, inherits the parent agent's tool set. When set to an
-    /// empty list, no tools are available to the subagent.
-    #[serde(default)]
-    pub tools: Option<Vec<String>>, // None => inherit; Some(vec) => allow-list
-}
-
-impl SubagentDefinition {
-    pub fn from_json_str(s: &str) -> Result<Self, serde_json::Error> {
-        serde_json::from_str::<Self>(s)
-    }
-
-    pub fn from_file(path: &Path) -> std::io::Result<Self> {
-        let contents = fs::read_to_string(path)?;
-        // Surface JSON parsing error with file context
-        serde_json::from_str::<Self>(&contents).map_err(|e| {
-            std::io::Error::new(
-                std::io::ErrorKind::InvalidData,
-                format!("invalid subagent JSON at {}: {e}", path.display()),
-            )
-        })
-    }
-}

codex-rs/core/src/subagents/mod.rs

@@ -1,6 +0,0 @@
-pub mod definition;
-pub mod registry;
-pub mod runner;
-pub mod tool;
-
-pub(crate) use tool::{SUBAGENT_LIST_TOOL, SUBAGENT_TOOL};

codex-rs/core/src/subagents/registry.rs

@@ -1,92 +0,0 @@
-use super::definition::SubagentDefinition;
-use std::collections::HashMap;
-use std::fs;
-use std::path::Path;
-use std::path::PathBuf;
-
-#[derive(Debug, Default, Clone)]
-pub struct SubagentRegistry {
-    /// Directory under the project (cwd/.codex/agents).
-    project_dir: Option<PathBuf>,
-    /// Directory under CODEX_HOME (~/.codex/agents).
-    user_dir: Option<PathBuf>,
-    /// Merged map: project definitions override user ones.
-    map: HashMap<String, SubagentDefinition>,
-}
-
-impl SubagentRegistry {
-    pub fn new(project_dir: Option<PathBuf>, user_dir: Option<PathBuf>) -> Self {
-        Self {
-            project_dir,
-            user_dir,
-            map: HashMap::new(),
-        }
-    }
-
-    /// Loads JSON files from user_dir then project_dir (project wins on conflict).
-    pub fn load(&mut self) {
-        let mut map: HashMap<String, SubagentDefinition> = HashMap::new();
-
-        // Load user definitions first
-        if let Some(dir) = &self.user_dir {
-            Self::load_from_dir_into(dir, &mut map);
-        }
-        // Then load project definitions which override on conflicts
-        if let Some(dir) = &self.project_dir {
-            Self::load_from_dir_into(dir, &mut map);
-        }
-
-        // Ensure a simple built‑in test subagent exists to validate wiring end‑to‑end.
-        // Users can override this by providing their own definition named "hello".
-        if !map.contains_key("hello") {
-            map.insert(
-                "hello".to_string(),
-                SubagentDefinition {
-                    name: "hello".to_string(),
-                    description: "Built‑in test subagent that replies with a greeting".to_string(),
-                    // Keep instructions narrow so models reliably output the intended text.
-                    instructions:
-                        "Reply with exactly this text and nothing else: Hello from subagent"
-                            .to_string(),
-                    // Disallow tool usage for the hello subagent.
-                    tools: Some(Vec::new()),
-                },
-            );
-        }
-
-        self.map = map;
-    }
-
-    pub fn get(&self, name: &str) -> Option<&SubagentDefinition> {
-        self.map.get(name)
-    }
-
-    pub fn all_names(&self) -> Vec<String> {
-        self.map.keys().cloned().collect()
-    }
-
-    fn load_from_dir_into(dir: &Path, out: &mut HashMap<String, SubagentDefinition>) {
-        let Ok(iter) = fs::read_dir(dir) else {
-            return;
-        };
-        for entry in iter.flatten() {
-            let path = entry.path();
-            if path.is_file()
-                && path
-                    .extension()
-                    .and_then(|e| e.to_str())
-                    .map(|e| e.eq_ignore_ascii_case("json"))
-                    .unwrap_or(false)
-            {
-                match SubagentDefinition::from_file(&path) {
-                    Ok(def) => {
-                        out.insert(def.name.clone(), def);
-                    }
-                    Err(e) => {
-                        tracing::warn!("Failed to load subagent from {}: {}", path.display(), e);
-                    }
-                }
-            }
-        }
-    }
-}

codex-rs/core/src/subagents/runner.rs

@@ -1,142 +0,0 @@
-use crate::codex::Codex;
-use crate::error::Result as CodexResult;
-
-use super::definition::SubagentDefinition;
-use super::registry::SubagentRegistry;
-
-/// Arguments expected for the `subagent.run` tool.
-#[derive(serde::Deserialize)]
-pub struct RunSubagentArgs {
-    pub name: String,
-    pub input: String,
-    #[serde(default)]
-    pub context: Option<String>,
-}
-
-/// Run a subagent in a nested Codex session and return the final message.
-pub(crate) async fn run(
-    sess: &crate::codex::Session,
-    turn_context: &crate::codex::TurnContext,
-    registry: &SubagentRegistry,
-    args: RunSubagentArgs,
-    _parent_sub_id: &str,
-) -> CodexResult<String> {
-    let def: &SubagentDefinition = registry.get(&args.name).ok_or_else(|| {
-        crate::error::CodexErr::Stream(format!("unknown subagent: {}", args.name), None)
-    })?;
-
-    let mut nested_cfg = (*sess.base_config()).clone();
-    nested_cfg.base_instructions = Some(def.instructions.clone());
-    nested_cfg.user_instructions = None;
-    nested_cfg.approval_policy = turn_context.approval_policy;
-    nested_cfg.sandbox_policy = turn_context.sandbox_policy.clone();
-    nested_cfg.cwd = turn_context.cwd.clone();
-    nested_cfg.include_subagent_tool = false;
-
-    let nested = Codex::spawn(nested_cfg, sess.auth_manager(), None).await?;
-    let nested_codex = nested.codex;
-
-    let subagent_id = uuid::Uuid::new_v4().to_string();
-    forward_begin(sess, _parent_sub_id, &subagent_id, &def.name).await;
-
-    let text = match args.context {
-        Some(ctx) if !ctx.trim().is_empty() => format!("{ctx}\n\n{input}", input = args.input),
-        _ => args.input,
-    };
-
-    nested_codex
-        .submit(crate::protocol::Op::UserInput {
-            items: vec![crate::protocol::InputItem::Text { text }],
-        })
-        .await
-        .map_err(|e| {
-            crate::error::CodexErr::Stream(format!("failed to submit to subagent: {e}"), None)
-        })?;
-
-    let mut last_message: Option<String> = None;
-    loop {
-        let ev = nested_codex.next_event().await?;
-        match ev.msg.clone() {
-            crate::protocol::EventMsg::AgentMessage(m) => {
-                last_message = Some(m.message);
-            }
-            crate::protocol::EventMsg::TaskComplete(t) => {
-                let _ = nested_codex.submit(crate::protocol::Op::Shutdown).await;
-                forward_forwarded(sess, _parent_sub_id, &subagent_id, &def.name, ev.msg).await;
-                forward_end(
-                    sess,
-                    _parent_sub_id,
-                    &subagent_id,
-                    &def.name,
-                    true,
-                    t.last_agent_message.clone(),
-                )
-                .await;
-                return Ok(t
-                    .last_agent_message
-                    .unwrap_or_else(|| last_message.unwrap_or_default()));
-            }
-            _ => {}
-        }
-        forward_forwarded(sess, _parent_sub_id, &subagent_id, &def.name, ev.msg).await;
-    }
-}
-
-async fn forward_begin(
-    sess: &crate::codex::Session,
-    parent_sub_id: &str,
-    subagent_id: &str,
-    name: &str,
-) {
-    sess
-        .send_event(crate::protocol::Event {
-            id: parent_sub_id.to_string(),
-            msg: crate::protocol::EventMsg::SubagentBegin(crate::protocol::SubagentBeginEvent {
-                subagent_id: subagent_id.to_string(),
-                name: name.to_string(),
-            }),
-        })
-        .await;
-}
-
-async fn forward_forwarded(
-    sess: &crate::codex::Session,
-    parent_sub_id: &str,
-    subagent_id: &str,
-    name: &str,
-    msg: crate::protocol::EventMsg,
-) {
-    sess
-        .send_event(crate::protocol::Event {
-            id: parent_sub_id.to_string(),
-            msg: crate::protocol::EventMsg::SubagentForwarded(
-                crate::protocol::SubagentForwardedEvent {
-                    subagent_id: subagent_id.to_string(),
-                    name: name.to_string(),
-                    event: Box::new(msg),
-                },
-            ),
-        })
-        .await;
-}
-
-async fn forward_end(
-    sess: &crate::codex::Session,
-    parent_sub_id: &str,
-    subagent_id: &str,
-    name: &str,
-    success: bool,
-    last_agent_message: Option<String>,
-) {
-    sess
-        .send_event(crate::protocol::Event {
-            id: parent_sub_id.to_string(),
-            msg: crate::protocol::EventMsg::SubagentEnd(crate::protocol::SubagentEndEvent {
-                subagent_id: subagent_id.to_string(),
-                name: name.to_string(),
-                success,
-                last_agent_message,
-            }),
-        })
-        .await;
-}

codex-rs/core/src/subagents/tool.rs

@@ -1,54 +0,0 @@
-use std::collections::BTreeMap;
-use std::sync::LazyLock;
-
-use crate::openai_tools::JsonSchema;
-use crate::openai_tools::OpenAiTool;
-use crate::openai_tools::ResponsesApiTool;
-
-pub(crate) static SUBAGENT_TOOL: LazyLock<OpenAiTool> = LazyLock::new(|| {
-    let mut properties = BTreeMap::new();
-    properties.insert(
-        "name".to_string(),
-        JsonSchema::String {
-            description: Some("Registered subagent name".to_string()),
-        },
-    );
-    properties.insert(
-        "input".to_string(),
-        JsonSchema::String {
-            description: Some("Task or instruction for the subagent".to_string()),
-        },
-    );
-    properties.insert(
-        "context".to_string(),
-        JsonSchema::String {
-            description: Some("Optional extra context to aid the task".to_string()),
-        },
-    );
-
-    OpenAiTool::Function(ResponsesApiTool {
-        name: "subagent_run".to_string(),
-        description: "Invoke a named subagent with isolated context and return its result"
-            .to_string(),
-        strict: false,
-        parameters: JsonSchema::Object {
-            properties,
-            required: Some(vec!["name".to_string(), "input".to_string()]),
-            additional_properties: Some(false),
-        },
-    })
-});
-
-pub(crate) static SUBAGENT_LIST_TOOL: LazyLock<OpenAiTool> = LazyLock::new(|| {
-    let properties = BTreeMap::new();
-    OpenAiTool::Function(ResponsesApiTool {
-        name: "subagent_list".to_string(),
-        description: "List available subagents (name and description). Call before subagent_run if unsure.".to_string(),
-        strict: false,
-        parameters: JsonSchema::Object {
-            properties,
-            required: None,
-            additional_properties: Some(false),
-        },
-    })
-});

codex-rs/core/tests/all.rs

@@ -0,0 +1,3 @@
+// Single integration test binary that aggregates all test modules.
+// The submodules live in `tests/all/`.
+mod suite;

codex-rs/core/tests/suite/mod.rs

@@ -0,0 +1,12 @@
+// Aggregates all former standalone integration tests as modules.
+
+mod cli_stream;
+mod client;
+mod compact;
+mod exec;
+mod exec_stream_events;
+mod live_cli;
+mod prompt_caching;
+mod seatbelt;
+mod stream_error_allows_next_turn;
+mod stream_no_completed;

codex-rs/core/tests/suite/prompt_caching.rs

@@ -107,8 +107,8 @@ async fn codex_mini_latest_tools() {
     assert_eq!(requests.len(), 2, "expected two POST requests");
 
     let expected_instructions = [
-        include_str!("../prompt.md"),
-        include_str!("../../apply-patch/apply_patch_tool_instructions.md"),
+        include_str!("../../prompt.md"),
+        include_str!("../../../apply-patch/apply_patch_tool_instructions.md"),
     ]
     .join("\n");
 
@@ -188,7 +188,7 @@ async fn prompt_tools_are_consistent_across_requests() {
     let requests = server.received_requests().await.unwrap();
     assert_eq!(requests.len(), 2, "expected two POST requests");
 
-    let expected_instructions: &str = include_str!("../prompt.md");
+    let expected_instructions: &str = include_str!("../../prompt.md");
     // our internal implementation is responsible for keeping tools in sync
     // with the OpenAI schema, so we just verify the tool presence here
     let expected_tools_names: &[&str] = &["shell", "update_plan", "apply_patch"];

codex-rs/exec/src/event_processor_with_human_output.rs

@@ -24,6 +24,7 @@ use codex_core::protocol::StreamErrorEvent;
 use codex_core::protocol::TaskCompleteEvent;
 use codex_core::protocol::TurnAbortReason;
 use codex_core::protocol::TurnDiffEvent;
+use codex_core::protocol::WebSearchBeginEvent;
 use owo_colors::OwoColorize;
 use owo_colors::Style;
 use shlex::try_join;
@@ -168,15 +169,6 @@ impl EventProcessor for EventProcessorWithHumanOutput {
     fn process_event(&mut self, event: Event) -> CodexStatus {
         let Event { id: _, msg } = event;
         match msg {
-            EventMsg::SubagentBegin(_) => {
-                // Ignore in human output for now.
-            }
-            EventMsg::SubagentForwarded(_) => {
-                // Ignore; TUI will render forwarded events.
-            }
-            EventMsg::SubagentEnd(_) => {
-                // Ignore in human output for now.
-            }
             EventMsg::Error(ErrorEvent { message }) => {
                 let prefix = "ERROR:".style(self.red);
                 ts_println!(self, "{prefix} {message}");
@@ -370,6 +362,9 @@ impl EventProcessor for EventProcessorWithHumanOutput {
                     }
                 }
             }
+            EventMsg::WebSearchBegin(WebSearchBeginEvent { call_id: _, query }) => {
+                ts_println!(self, "🌐 {query}");
+            }
             EventMsg::PatchApplyBegin(PatchApplyBeginEvent {
                 call_id,
                 auto_approved,

codex-rs/exec/src/event_processor_with_json_output.rs

@@ -41,12 +41,6 @@ impl EventProcessor for EventProcessorWithJsonOutput {
 
     fn process_event(&mut self, event: Event) -> CodexStatus {
         match event.msg {
-            EventMsg::SubagentBegin(_)
-            | EventMsg::SubagentForwarded(_)
-            | EventMsg::SubagentEnd(_) => {
-                // Ignored for JSON output in exec for now.
-                CodexStatus::Running
-            }
             EventMsg::AgentMessageDelta(_) | EventMsg::AgentReasoningDelta(_) => {
                 // Suppress streaming events in JSON mode.
                 CodexStatus::Running

codex-rs/exec/src/lib.rs

@@ -146,11 +146,11 @@ pub async fn run_main(cli: Cli, codex_linux_sandbox_exe: Option<PathBuf>) -> any
         model_provider,
         codex_linux_sandbox_exe,
         base_instructions: None,
-        include_subagent_tool: None,
         include_plan_tool: None,
         include_apply_patch_tool: None,
         disable_response_storage: oss.then_some(true),
         show_raw_agent_reasoning: oss.then_some(true),
+        tools_web_search_request: None,
     };
     // Parse `-c` overrides.
     let cli_kv_overrides = match config_overrides.parse_overrides() {

codex-rs/exec/tests/all.rs

@@ -0,0 +1,3 @@
+// Single integration test binary that aggregates all test modules.
+// The submodules live in `tests/suite/`.
+mod suite;

codex-rs/exec/tests/apply_patch.rs

@@ -1,339 +0,0 @@
-#![allow(clippy::expect_used, clippy::unwrap_used)]
-
-use anyhow::Context;
-use assert_cmd::prelude::*;
-use codex_core::CODEX_APPLY_PATCH_ARG1;
-use std::fs;
-use std::process::Command;
-use tempfile::tempdir;
-
-/// While we may add an `apply-patch` subcommand to the `codex` CLI multitool
-/// at some point, we must ensure that the smaller `codex-exec` CLI can still
-/// emulate the `apply_patch` CLI.
-#[test]
-fn test_standalone_exec_cli_can_use_apply_patch() -> anyhow::Result<()> {
-    let tmp = tempdir()?;
-    let relative_path = "source.txt";
-    let absolute_path = tmp.path().join(relative_path);
-    fs::write(&absolute_path, "original content\n")?;
-
-    Command::cargo_bin("codex-exec")
-        .context("should find binary for codex-exec")?
-        .arg(CODEX_APPLY_PATCH_ARG1)
-        .arg(
-            r#"*** Begin Patch
-*** Update File: source.txt
-@@
--original content
-+modified by apply_patch
-*** End Patch"#,
-        )
-        .current_dir(tmp.path())
-        .assert()
-        .success()
-        .stdout("Success. Updated the following files:\nM source.txt\n")
-        .stderr(predicates::str::is_empty());
-    assert_eq!(
-        fs::read_to_string(absolute_path)?,
-        "modified by apply_patch\n"
-    );
-    Ok(())
-}
-
-#[cfg(not(target_os = "windows"))]
-#[tokio::test]
-async fn test_apply_patch_tool() -> anyhow::Result<()> {
-    use core_test_support::load_sse_fixture_with_id_from_str;
-    use tempfile::TempDir;
-    use wiremock::Mock;
-    use wiremock::MockServer;
-    use wiremock::ResponseTemplate;
-    use wiremock::matchers::method;
-    use wiremock::matchers::path;
-
-    const SSE_TOOL_CALL_ADD: &str = r#"[
-  {
-    "type": "response.output_item.done",
-    "item": {
-      "type": "function_call",
-      "name": "apply_patch",
-      "arguments": "{\n  \"input\": \"*** Begin Patch\\n*** Add File: test.md\\n+Hello world\\n*** End Patch\"\n}",
-      "call_id": "__ID__"
-    }
-  },
-  {
-    "type": "response.completed",
-    "response": {
-      "id": "__ID__",
-      "usage": {
-        "input_tokens": 0,
-        "input_tokens_details": null,
-        "output_tokens": 0,
-        "output_tokens_details": null,
-        "total_tokens": 0
-      },
-      "output": []
-    }
-  }
-]"#;
-
-    const SSE_TOOL_CALL_UPDATE: &str = r#"[
-  {
-    "type": "response.output_item.done",
-    "item": {
-      "type": "function_call",
-      "name": "apply_patch",
-      "arguments": "{\n  \"input\": \"*** Begin Patch\\n*** Update File: test.md\\n@@\\n-Hello world\\n+Final text\\n*** End Patch\"\n}",
-      "call_id": "__ID__"
-    }
-  },
-  {
-    "type": "response.completed",
-    "response": {
-      "id": "__ID__",
-      "usage": {
-        "input_tokens": 0,
-        "input_tokens_details": null,
-        "output_tokens": 0,
-        "output_tokens_details": null,
-        "total_tokens": 0
-      },
-      "output": []
-    }
-  }
-]"#;
-
-    const SSE_TOOL_CALL_COMPLETED: &str = r#"[
-  {
-    "type": "response.completed",
-    "response": {
-      "id": "__ID__",
-      "usage": {
-        "input_tokens": 0,
-        "input_tokens_details": null,
-        "output_tokens": 0,
-        "output_tokens_details": null,
-        "total_tokens": 0
-      },
-      "output": []
-    }
-  }
-]"#;
-
-    // Start a mock model server
-    let server = MockServer::start().await;
-
-    // First response: model calls apply_patch to create test.md
-    let first = ResponseTemplate::new(200)
-        .insert_header("content-type", "text/event-stream")
-        .set_body_raw(
-            load_sse_fixture_with_id_from_str(SSE_TOOL_CALL_ADD, "call1"),
-            "text/event-stream",
-        );
-
-    Mock::given(method("POST"))
-        // .and(path("/v1/responses"))
-        .respond_with(first)
-        .up_to_n_times(1)
-        .mount(&server)
-        .await;
-
-    // Second response: model calls apply_patch to update test.md
-    let second = ResponseTemplate::new(200)
-        .insert_header("content-type", "text/event-stream")
-        .set_body_raw(
-            load_sse_fixture_with_id_from_str(SSE_TOOL_CALL_UPDATE, "call2"),
-            "text/event-stream",
-        );
-
-    Mock::given(method("POST"))
-        .and(path("/v1/responses"))
-        .respond_with(second)
-        .up_to_n_times(1)
-        .mount(&server)
-        .await;
-
-    let final_completed = ResponseTemplate::new(200)
-        .insert_header("content-type", "text/event-stream")
-        .set_body_raw(
-            load_sse_fixture_with_id_from_str(SSE_TOOL_CALL_COMPLETED, "resp3"),
-            "text/event-stream",
-        );
-
-    Mock::given(method("POST"))
-        .and(path("/v1/responses"))
-        .respond_with(final_completed)
-        .expect(1)
-        .mount(&server)
-        .await;
-
-    let tmp_cwd = TempDir::new().unwrap();
-    Command::cargo_bin("codex-exec")
-        .context("should find binary for codex-exec")?
-        .current_dir(tmp_cwd.path())
-        .env("CODEX_HOME", tmp_cwd.path())
-        .env("OPENAI_API_KEY", "dummy")
-        .env("OPENAI_BASE_URL", format!("{}/v1", server.uri()))
-        .arg("--skip-git-repo-check")
-        .arg("-s")
-        .arg("workspace-write")
-        .arg("foo")
-        .assert()
-        .success();
-
-    // Verify final file contents
-    let final_path = tmp_cwd.path().join("test.md");
-    let contents = std::fs::read_to_string(&final_path)
-        .unwrap_or_else(|e| panic!("failed reading {}: {e}", final_path.display()));
-    assert_eq!(contents, "Final text\n");
-    Ok(())
-}
-
-#[cfg(not(target_os = "windows"))]
-#[tokio::test]
-async fn test_apply_patch_freeform_tool() -> anyhow::Result<()> {
-    use core_test_support::load_sse_fixture_with_id_from_str;
-    use tempfile::TempDir;
-    use wiremock::Mock;
-    use wiremock::MockServer;
-    use wiremock::ResponseTemplate;
-    use wiremock::matchers::method;
-    use wiremock::matchers::path;
-
-    const SSE_TOOL_CALL_ADD: &str = r#"[
-  {
-    "type": "response.output_item.done",
-    "item": {
-      "type": "custom_tool_call",
-      "name": "apply_patch",
-      "input": "*** Begin Patch\n*** Add File: test.md\n+Hello world\n*** End Patch",
-      "call_id": "__ID__"
-    }
-  },
-  {
-    "type": "response.completed",
-    "response": {
-      "id": "__ID__",
-      "usage": {
-        "input_tokens": 0,
-        "input_tokens_details": null,
-        "output_tokens": 0,
-        "output_tokens_details": null,
-        "total_tokens": 0
-      },
-      "output": []
-    }
-  }
-]"#;
-
-    const SSE_TOOL_CALL_UPDATE: &str = r#"[
-  {
-    "type": "response.output_item.done",
-    "item": {
-      "type": "custom_tool_call",
-      "name": "apply_patch",
-      "input": "*** Begin Patch\n*** Update File: test.md\n@@\n-Hello world\n+Final text\n*** End Patch",
-      "call_id": "__ID__"
-    }
-  },
-  {
-    "type": "response.completed",
-    "response": {
-      "id": "__ID__",
-      "usage": {
-        "input_tokens": 0,
-        "input_tokens_details": null,
-        "output_tokens": 0,
-        "output_tokens_details": null,
-        "total_tokens": 0
-      },
-      "output": []
-    }
-  }
-]"#;
-
-    const SSE_TOOL_CALL_COMPLETED: &str = r#"[
-  {
-    "type": "response.completed",
-    "response": {
-      "id": "__ID__",
-      "usage": {
-        "input_tokens": 0,
-        "input_tokens_details": null,
-        "output_tokens": 0,
-        "output_tokens_details": null,
-        "total_tokens": 0
-      },
-      "output": []
-    }
-  }
-]"#;
-
-    // Start a mock model server
-    let server = MockServer::start().await;
-
-    // First response: model calls apply_patch to create test.md
-    let first = ResponseTemplate::new(200)
-        .insert_header("content-type", "text/event-stream")
-        .set_body_raw(
-            load_sse_fixture_with_id_from_str(SSE_TOOL_CALL_ADD, "call1"),
-            "text/event-stream",
-        );
-
-    Mock::given(method("POST"))
-        // .and(path("/v1/responses"))
-        .respond_with(first)
-        .up_to_n_times(1)
-        .mount(&server)
-        .await;
-
-    // Second response: model calls apply_patch to update test.md
-    let second = ResponseTemplate::new(200)
-        .insert_header("content-type", "text/event-stream")
-        .set_body_raw(
-            load_sse_fixture_with_id_from_str(SSE_TOOL_CALL_UPDATE, "call2"),
-            "text/event-stream",
-        );
-
-    Mock::given(method("POST"))
-        .and(path("/v1/responses"))
-        .respond_with(second)
-        .up_to_n_times(1)
-        .mount(&server)
-        .await;
-
-    let final_completed = ResponseTemplate::new(200)
-        .insert_header("content-type", "text/event-stream")
-        .set_body_raw(
-            load_sse_fixture_with_id_from_str(SSE_TOOL_CALL_COMPLETED, "resp3"),
-            "text/event-stream",
-        );
-
-    Mock::given(method("POST"))
-        // .and(path("/v1/responses"))
-        .respond_with(final_completed)
-        .expect(1)
-        .mount(&server)
-        .await;
-
-    let tmp_cwd = TempDir::new().unwrap();
-    Command::cargo_bin("codex-exec")
-        .context("should find binary for codex-exec")?
-        .current_dir(tmp_cwd.path())
-        .env("CODEX_HOME", tmp_cwd.path())
-        .env("OPENAI_API_KEY", "dummy")
-        .env("OPENAI_BASE_URL", format!("{}/v1", server.uri()))
-        .arg("--skip-git-repo-check")
-        .arg("-s")
-        .arg("workspace-write")
-        .arg("foo")
-        .assert()
-        .success();
-
-    // Verify final file contents
-    let final_path = tmp_cwd.path().join("test.md");
-    let contents = std::fs::read_to_string(&final_path)
-        .unwrap_or_else(|e| panic!("failed reading {}: {e}", final_path.display()));
-    assert_eq!(contents, "Final text\n");
-    Ok(())
-}

codex-rs/exec/tests/fixtures/apply_patch_freeform_final.txt

@@ -0,0 +1,4 @@
+class BaseClass:
+  def method():
+
+    return True

codex-rs/exec/tests/fixtures/sse_apply_patch_add.json

@@ -0,0 +1,25 @@
+[
+  {
+    "type": "response.output_item.done",
+    "item": {
+      "type": "custom_tool_call",
+      "name": "apply_patch",
+      "input": "*** Begin Patch\n*** Add File: test.md\n+Hello world\n*** End Patch",
+      "call_id": "__ID__"
+    }
+  },
+  {
+    "type": "response.completed",
+    "response": {
+      "id": "__ID__",
+      "usage": {
+        "input_tokens": 0,
+        "input_tokens_details": null,
+        "output_tokens": 0,
+        "output_tokens_details": null,
+        "total_tokens": 0
+      },
+      "output": []
+    }
+  }
+]

codex-rs/exec/tests/fixtures/sse_apply_patch_context_update.json

@@ -0,0 +1,25 @@
+[
+  {
+    "type": "response.output_item.done",
+    "item": {
+      "type": "custom_tool_call",
+      "name": "apply_patch",
+      "input": "*** Begin Patch\n*** Update File: app.py\n@@ class BaseClass:\n@@   def method():\n-    return False\n+    return True\n*** End Patch",
+      "call_id": "__ID__"
+    }
+  },
+  {
+    "type": "response.completed",
+    "response": {
+      "id": "__ID__",
+      "usage": {
+        "input_tokens": 0,
+        "input_tokens_details": null,
+        "output_tokens": 0,
+        "output_tokens_details": null,
+        "total_tokens": 0
+      },
+      "output": []
+    }
+  }
+]

codex-rs/exec/tests/fixtures/sse_apply_patch_freeform_add.json

@@ -0,0 +1,25 @@
+[
+  {
+    "type": "response.output_item.done",
+    "item": {
+      "type": "custom_tool_call",
+      "name": "apply_patch",
+      "input": "*** Begin Patch\n*** Add File: app.py\n+class BaseClass:\n+  def method():\n+    return False\n*** End Patch",
+      "call_id": "__ID__"
+    }
+  },
+  {
+    "type": "response.completed",
+    "response": {
+      "id": "__ID__",
+      "usage": {
+        "input_tokens": 0,
+        "input_tokens_details": null,
+        "output_tokens": 0,
+        "output_tokens_details": null,
+        "total_tokens": 0
+      },
+      "output": []
+    }
+  }
+]

codex-rs/exec/tests/fixtures/sse_apply_patch_freeform_update.json

@@ -0,0 +1,25 @@
+[
+  {
+    "type": "response.output_item.done",
+    "item": {
+      "type": "custom_tool_call",
+      "name": "apply_patch",
+      "input": "*** Begin Patch\n*** Update File: app.py\n@@  def method():\n-    return False\n+\n+    return True\n*** End Patch",
+      "call_id": "__ID__"
+    }
+  },
+  {
+    "type": "response.completed",
+    "response": {
+      "id": "__ID__",
+      "usage": {
+        "input_tokens": 0,
+        "input_tokens_details": null,
+        "output_tokens": 0,
+        "output_tokens_details": null,
+        "total_tokens": 0
+      },
+      "output": []
+    }
+  }
+]

codex-rs/exec/tests/fixtures/sse_apply_patch_update.json

@@ -0,0 +1,25 @@
+[
+  {
+    "type": "response.output_item.done",
+    "item": {
+      "type": "function_call",
+      "name": "apply_patch",
+      "arguments": "{\n  \"input\": \"*** Begin Patch\\n*** Update File: test.md\\n@@\\n-Hello world\\n+Final text\\n*** End Patch\"\n}",
+      "call_id": "__ID__"
+    }
+  },
+  {
+    "type": "response.completed",
+    "response": {
+      "id": "__ID__",
+      "usage": {
+        "input_tokens": 0,
+        "input_tokens_details": null,
+        "output_tokens": 0,
+        "output_tokens_details": null,
+        "total_tokens": 0
+      },
+      "output": []
+    }
+  }
+]

codex-rs/exec/tests/fixtures/sse_response_completed.json

@@ -0,0 +1,16 @@
+[
+  {
+    "type": "response.completed",
+    "response": {
+      "id": "__ID__",
+      "usage": {
+        "input_tokens": 0,
+        "input_tokens_details": null,
+        "output_tokens": 0,
+        "output_tokens_details": null,
+        "total_tokens": 0
+      },
+      "output": []
+    }
+  }
+]

codex-rs/exec/tests/suite/apply_patch.rs

@@ -0,0 +1,146 @@
+#![allow(clippy::expect_used, clippy::unwrap_used)]
+
+use anyhow::Context;
+use assert_cmd::prelude::*;
+use codex_core::CODEX_APPLY_PATCH_ARG1;
+use std::fs;
+use std::process::Command;
+use tempfile::tempdir;
+
+/// While we may add an `apply-patch` subcommand to the `codex` CLI multitool
+/// at some point, we must ensure that the smaller `codex-exec` CLI can still
+/// emulate the `apply_patch` CLI.
+#[test]
+fn test_standalone_exec_cli_can_use_apply_patch() -> anyhow::Result<()> {
+    let tmp = tempdir()?;
+    let relative_path = "source.txt";
+    let absolute_path = tmp.path().join(relative_path);
+    fs::write(&absolute_path, "original content\n")?;
+
+    Command::cargo_bin("codex-exec")
+        .context("should find binary for codex-exec")?
+        .arg(CODEX_APPLY_PATCH_ARG1)
+        .arg(
+            r#"*** Begin Patch
+*** Update File: source.txt
+@@
+-original content
++modified by apply_patch
+*** End Patch"#,
+        )
+        .current_dir(tmp.path())
+        .assert()
+        .success()
+        .stdout("Success. Updated the following files:\nM source.txt\n")
+        .stderr(predicates::str::is_empty());
+    assert_eq!(
+        fs::read_to_string(absolute_path)?,
+        "modified by apply_patch\n"
+    );
+    Ok(())
+}
+
+#[cfg(not(target_os = "windows"))]
+#[tokio::test(flavor = "multi_thread", worker_threads = 4)]
+async fn test_apply_patch_tool() -> anyhow::Result<()> {
+    use crate::suite::common::run_e2e_exec_test;
+    use codex_core::spawn::CODEX_SANDBOX_NETWORK_DISABLED_ENV_VAR;
+
+    if std::env::var(CODEX_SANDBOX_NETWORK_DISABLED_ENV_VAR).is_ok() {
+        println!(
+            "Skipping test because it cannot execute when network is disabled in a Codex sandbox."
+        );
+        return Ok(());
+    }
+
+    let tmp_cwd = tempdir().expect("failed to create temp dir");
+    let tmp_path = tmp_cwd.path().to_path_buf();
+    run_e2e_exec_test(
+        tmp_cwd.path(),
+        vec![
+            include_str!("../fixtures/sse_apply_patch_add.json").to_string(),
+            include_str!("../fixtures/sse_apply_patch_update.json").to_string(),
+            include_str!("../fixtures/sse_response_completed.json").to_string(),
+        ],
+    )
+    .await;
+
+    let final_path = tmp_path.join("test.md");
+    let contents = std::fs::read_to_string(&final_path)
+        .unwrap_or_else(|e| panic!("failed reading {}: {e}", final_path.display()));
+    assert_eq!(contents, "Final text\n");
+    Ok(())
+}
+
+#[cfg(not(target_os = "windows"))]
+#[tokio::test(flavor = "multi_thread", worker_threads = 4)]
+async fn test_apply_patch_freeform_tool() -> anyhow::Result<()> {
+    use crate::suite::common::run_e2e_exec_test;
+    use codex_core::spawn::CODEX_SANDBOX_NETWORK_DISABLED_ENV_VAR;
+
+    if std::env::var(CODEX_SANDBOX_NETWORK_DISABLED_ENV_VAR).is_ok() {
+        println!(
+            "Skipping test because it cannot execute when network is disabled in a Codex sandbox."
+        );
+        return Ok(());
+    }
+
+    let tmp_cwd = tempdir().expect("failed to create temp dir");
+    run_e2e_exec_test(
+        tmp_cwd.path(),
+        vec![
+            include_str!("../fixtures/sse_apply_patch_freeform_add.json").to_string(),
+            include_str!("../fixtures/sse_apply_patch_freeform_update.json").to_string(),
+            include_str!("../fixtures/sse_response_completed.json").to_string(),
+        ],
+    )
+    .await;
+
+    // Verify final file contents
+    let final_path = tmp_cwd.path().join("app.py");
+    let contents = std::fs::read_to_string(&final_path)
+        .unwrap_or_else(|e| panic!("failed reading {}: {e}", final_path.display()));
+    assert_eq!(
+        contents,
+        include_str!("../fixtures/apply_patch_freeform_final.txt")
+    );
+    Ok(())
+}
+
+#[cfg(not(target_os = "windows"))]
+#[tokio::test]
+async fn test_apply_patch_context() -> anyhow::Result<()> {
+    use crate::suite::common::run_e2e_exec_test;
+    use codex_core::spawn::CODEX_SANDBOX_NETWORK_DISABLED_ENV_VAR;
+
+    if std::env::var(CODEX_SANDBOX_NETWORK_DISABLED_ENV_VAR).is_ok() {
+        println!(
+            "Skipping test because it cannot execute when network is disabled in a Codex sandbox."
+        );
+        return Ok(());
+    }
+
+    let tmp_cwd = tempdir().expect("failed to create temp dir");
+    run_e2e_exec_test(
+        tmp_cwd.path(),
+        vec![
+            include_str!("../fixtures/sse_apply_patch_freeform_add.json").to_string(),
+            include_str!("../fixtures/sse_apply_patch_context_update.json").to_string(),
+            include_str!("../fixtures/sse_response_completed.json").to_string(),
+        ],
+    )
+    .await;
+
+    // Verify final file contents
+    let final_path = tmp_cwd.path().join("app.py");
+    let contents = std::fs::read_to_string(&final_path)
+        .unwrap_or_else(|e| panic!("failed reading {}: {e}", final_path.display()));
+    assert_eq!(
+        contents,
+        r#"class BaseClass:
+  def method():
+    return True
+"#
+    );
+    Ok(())
+}

codex-rs/exec/tests/suite/common.rs

@@ -0,0 +1,73 @@
+// this file is only used for e2e tests which are currently disabled on windows
+#![cfg(not(target_os = "windows"))]
+#![allow(clippy::expect_used)]
+
+use anyhow::Context;
+use assert_cmd::prelude::*;
+use core_test_support::load_sse_fixture_with_id_from_str;
+use std::path::Path;
+use std::process::Command;
+use std::sync::atomic::AtomicUsize;
+use std::sync::atomic::Ordering;
+use wiremock::Mock;
+use wiremock::MockServer;
+use wiremock::matchers::method;
+use wiremock::matchers::path;
+
+use wiremock::Respond;
+
+struct SeqResponder {
+    num_calls: AtomicUsize,
+    responses: Vec<String>,
+}
+
+impl Respond for SeqResponder {
+    fn respond(&self, _: &wiremock::Request) -> wiremock::ResponseTemplate {
+        let call_num = self.num_calls.fetch_add(1, Ordering::SeqCst);
+        match self.responses.get(call_num) {
+            Some(body) => wiremock::ResponseTemplate::new(200)
+                .insert_header("content-type", "text/event-stream")
+                .set_body_raw(
+                    load_sse_fixture_with_id_from_str(body, &format!("request_{}", call_num)),
+                    "text/event-stream",
+                ),
+            None => panic!("no response for {call_num}"),
+        }
+    }
+}
+
+/// Helper function to run an E2E test of a codex-exec call. Starts a wiremock
+/// server, and returns the response_streams in order for each api call. Runs
+/// the codex-exec command with the wiremock server as the model server.
+pub(crate) async fn run_e2e_exec_test(cwd: &Path, response_streams: Vec<String>) {
+    let server = MockServer::start().await;
+
+    let num_calls = response_streams.len();
+    let seq_responder = SeqResponder {
+        num_calls: AtomicUsize::new(0),
+        responses: response_streams,
+    };
+
+    Mock::given(method("POST"))
+        .and(path("/v1/responses"))
+        .respond_with(seq_responder)
+        .expect(num_calls as u64)
+        .mount(&server)
+        .await;
+
+    let cwd = cwd.to_path_buf();
+    let uri = server.uri();
+    Command::cargo_bin("codex-exec")
+        .context("should find binary for codex-exec")
+        .expect("should find binary for codex-exec")
+        .current_dir(cwd.clone())
+        .env("CODEX_HOME", cwd.clone())
+        .env("OPENAI_API_KEY", "dummy")
+        .env("OPENAI_BASE_URL", format!("{}/v1", uri))
+        .arg("--skip-git-repo-check")
+        .arg("-s")
+        .arg("danger-full-access")
+        .arg("foo")
+        .assert()
+        .success();
+}

codex-rs/exec/tests/suite/mod.rs

@@ -0,0 +1,4 @@
+// Aggregates all former standalone integration tests as modules.
+mod apply_patch;
+mod common;
+mod sandbox;

codex-rs/execpolicy/tests/all.rs

@@ -0,0 +1,3 @@
+// Single integration test binary that aggregates all test modules.
+// The submodules live in `tests/suite/`.
+mod suite;

codex-rs/execpolicy/tests/suite/mod.rs

@@ -0,0 +1,10 @@
+// Aggregates all former standalone integration tests as modules.
+mod bad;
+mod cp;
+mod good;
+mod head;
+mod literal;
+mod ls;
+mod parse_sed_command;
+mod pwd;
+mod sed;

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

@@ -0,0 +1,3 @@
+// Single integration test binary that aggregates all test modules.
+// The submodules live in `tests/suite/`.
+mod suite;

codex-rs/linux-sandbox/tests/suite/mod.rs

@@ -0,0 +1,2 @@
+// Aggregates all former standalone integration tests as modules.
+mod landlock;

codex-rs/login/tests/all.rs

@@ -0,0 +1,3 @@
+// Single integration test binary that aggregates all test modules.
+// The submodules live in `tests/suite/`.
+mod suite;

codex-rs/login/tests/suite/mod.rs

@@ -0,0 +1,2 @@
+// Aggregates all former standalone integration tests as modules.
+mod login_server_e2e;

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

@@ -736,9 +736,9 @@ fn derive_config_from_params(
         base_instructions,
         include_plan_tool,
         include_apply_patch_tool,
-        include_subagent_tool: None,
         disable_response_storage: None,
         show_raw_agent_reasoning: None,
+        tools_web_search_request: None,
     };
 
     let cli_overrides = cli_overrides

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

@@ -161,9 +161,9 @@ impl CodexToolCallParam {
             base_instructions,
             include_plan_tool,
             include_apply_patch_tool: None,
-            include_subagent_tool: None,
             disable_response_storage: None,
             show_raw_agent_reasoning: None,
+            tools_web_search_request: None,
         };
 
         let cli_overrides = cli_overrides

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

@@ -174,11 +174,6 @@ async fn run_codex_tool_session_inner(
                     .await;
 
                 match event.msg {
-                    EventMsg::SubagentBegin(_)
-                    | EventMsg::SubagentForwarded(_)
-                    | EventMsg::SubagentEnd(_) => {
-                        // Ignore subagent orchestration for MCP echoing.
-                    }
                     EventMsg::ExecApprovalRequest(ExecApprovalRequestEvent {
                         command,
                         cwd,
@@ -277,6 +272,7 @@ async fn run_codex_tool_session_inner(
                     | EventMsg::PatchApplyBegin(_)
                     | EventMsg::PatchApplyEnd(_)
                     | EventMsg::TurnDiff(_)
+                    | EventMsg::WebSearchBegin(_)
                     | EventMsg::GetHistoryEntryResponse(_)
                     | EventMsg::PlanUpdate(_)
                     | EventMsg::TurnAborted(_)

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

@@ -0,0 +1,3 @@
+// Single integration test binary that aggregates all test modules.
+// The submodules live in `tests/suite/`.
+mod suite;

codex-rs/mcp-server/tests/suite/mod.rs

@@ -0,0 +1,8 @@
+// Aggregates all former standalone integration tests as modules.
+mod auth;
+mod codex_message_processor_flow;
+mod codex_tool;
+mod create_conversation;
+mod interrupt;
+mod login;
+mod send_message;

codex-rs/mcp-types/tests/all.rs

@@ -0,0 +1,3 @@
+// Single integration test binary that aggregates all test modules.
+// The submodules live in `tests/suite/`.
+mod suite;

codex-rs/mcp-types/tests/suite/mod.rs

@@ -0,0 +1,3 @@
+// Aggregates all former standalone integration tests as modules.
+mod initialize;
+mod progress_notification;

codex-rs/protocol-ts/src/lib.rs

@@ -48,6 +48,8 @@ pub fn generate_ts(out_dir: &Path, prettier: Option<&Path>) -> Result<()> {
     codex_protocol::mcp_protocol::ExecCommandApprovalResponse::export_all_to(out_dir)?;
     codex_protocol::mcp_protocol::ServerNotification::export_all_to(out_dir)?;
 
+    generate_index_ts(out_dir)?;
+
     // Prepend header to each generated .ts file
     let ts_files = ts_files_in(out_dir)?;
     for file in &ts_files {
@@ -109,5 +111,39 @@ fn ts_files_in(dir: &Path) -> Result<Vec<PathBuf>> {
             files.push(path);
         }
     }
+    files.sort();
     Ok(files)
 }
+
+/// Generate an index.ts file that re-exports all generated types.
+/// This allows consumers to import all types from a single file.
+fn generate_index_ts(out_dir: &Path) -> Result<PathBuf> {
+    let mut entries: Vec<String> = Vec::new();
+    let mut stems: Vec<String> = ts_files_in(out_dir)?
+        .into_iter()
+        .filter_map(|p| {
+            let stem = p.file_stem()?.to_string_lossy().into_owned();
+            if stem == "index" { None } else { Some(stem) }
+        })
+        .collect();
+    stems.sort();
+    stems.dedup();
+
+    for name in stems {
+        entries.push(format!("export type {{ {name} }} from \"./{name}\";\n"));
+    }
+
+    let mut content =
+        String::with_capacity(HEADER.len() + entries.iter().map(|s| s.len()).sum::<usize>());
+    content.push_str(HEADER);
+    for line in &entries {
+        content.push_str(line);
+    }
+
+    let index_path = out_dir.join("index.ts");
+    let mut f = fs::File::create(&index_path)
+        .with_context(|| format!("Failed to create {}", index_path.display()))?;
+    f.write_all(content.as_bytes())
+        .with_context(|| format!("Failed to write {}", index_path.display()))?;
+    Ok(index_path)
+}

codex-rs/protocol/src/protocol.rs

@@ -437,6 +437,8 @@ pub enum EventMsg {
 
     McpToolCallEnd(McpToolCallEndEvent),
 
+    WebSearchBegin(WebSearchBeginEvent),
+
     /// Notification that the server is about to execute a command.
     ExecCommandBegin(ExecCommandBeginEvent),
 
@@ -478,14 +480,6 @@ pub enum EventMsg {
     ShutdownComplete,
 
     ConversationHistory(ConversationHistoryResponseEvent),
-
-    // --- Subagent orchestration events ---
-    /// Emitted when a subagent starts.
-    SubagentBegin(SubagentBeginEvent),
-    /// Forwards a nested event produced by a running subagent.
-    SubagentForwarded(SubagentForwardedEvent),
-    /// Emitted when a subagent finishes.
-    SubagentEnd(SubagentEndEvent),
 }
 
 // Individual event payload types matching each `EventMsg` variant.
@@ -509,28 +503,6 @@ pub struct TokenUsage {
     pub total_tokens: u64,
 }
 
-#[derive(Debug, Clone, Deserialize, Serialize)]
-pub struct SubagentBeginEvent {
-    pub subagent_id: String,
-    pub name: String,
-}
-
-#[derive(Debug, Clone, Deserialize, Serialize)]
-pub struct SubagentEndEvent {
-    pub subagent_id: String,
-    pub name: String,
-    pub success: bool,
-    #[serde(skip_serializing_if = "Option::is_none")]
-    pub last_agent_message: Option<String>,
-}
-
-#[derive(Debug, Clone, Deserialize, Serialize)]
-pub struct SubagentForwardedEvent {
-    pub subagent_id: String,
-    pub name: String,
-    pub event: Box<EventMsg>,
-}
-
 impl TokenUsage {
     pub fn is_zero(&self) -> bool {
         self.total_tokens == 0
@@ -688,6 +660,12 @@ impl McpToolCallEndEvent {
     }
 }
 
+#[derive(Debug, Clone, Deserialize, Serialize)]
+pub struct WebSearchBeginEvent {
+    pub call_id: String,
+    pub query: String,
+}
+
 /// Response payload for `Op::GetHistory` containing the current session's
 /// in-memory transcript.
 #[derive(Debug, Clone, Deserialize, Serialize)]