Enforce user input length cap (#12823)

Currently there is no bound on the length of a user message submitted in
the TUI or through the app server interface. That means users can paste
many megabytes of text, which can lead to bad performance, hangs, and
crashes. In extreme cases, it can lead to a [kernel
panic](https://github.com/openai/codex/issues/12323).

This PR limits the length of a user input to 2**20 (about 1M)
characters. This value was chosen because it fills the entire context
window on the latest models, so accepting longer inputs wouldn't make
sense anyway.

Summary
- add a shared `MAX_USER_INPUT_TEXT_CHARS` constant in codex-protocol
and surface it in TUI and app server code
- block oversized submissions in the TUI submit flow and emit error
history cells when validation fails
- reject heavy app-server requests with JSON-RPC `-32602` and structured
`input_too_large` data, plus document the behavior

Testing
- ran the IDE extension with this change and verified that when I
attempt to paste a user message that's several MB long, it correctly
reports an error instead of crashing or making my computer hot.

codex-rs/app-server/tests/suite/v2/turn_steer.rs

@@ -6,6 +6,8 @@ use app_test_support::create_mock_responses_server_sequence;
 use app_test_support::create_mock_responses_server_sequence_unchecked;
 use app_test_support::create_shell_command_sse_response;
 use app_test_support::to_response;
+use codex_app_server::INPUT_TOO_LARGE_ERROR_CODE;
+use codex_app_server::INVALID_PARAMS_ERROR_CODE;
 use codex_app_server_protocol::JSONRPCError;
 use codex_app_server_protocol::JSONRPCNotification;
 use codex_app_server_protocol::JSONRPCResponse;
@@ -17,6 +19,7 @@ use codex_app_server_protocol::TurnStartResponse;
 use codex_app_server_protocol::TurnSteerParams;
 use codex_app_server_protocol::TurnSteerResponse;
 use codex_app_server_protocol::UserInput as V2UserInput;
+use codex_protocol::user_input::MAX_USER_INPUT_TEXT_CHARS;
 use tempfile::TempDir;
 use tokio::time::timeout;
 
@@ -67,6 +70,109 @@ async fn turn_steer_requires_active_turn() -> Result<()> {
     Ok(())
 }
 
+#[tokio::test]
+async fn turn_steer_rejects_oversized_text_input() -> Result<()> {
+    #[cfg(target_os = "windows")]
+    let shell_command = vec![
+        "powershell".to_string(),
+        "-Command".to_string(),
+        "Start-Sleep -Seconds 10".to_string(),
+    ];
+    #[cfg(not(target_os = "windows"))]
+    let shell_command = vec!["sleep".to_string(), "10".to_string()];
+
+    let tmp = TempDir::new()?;
+    let codex_home = tmp.path().join("codex_home");
+    std::fs::create_dir(&codex_home)?;
+    let working_directory = tmp.path().join("workdir");
+    std::fs::create_dir(&working_directory)?;
+
+    let server =
+        create_mock_responses_server_sequence_unchecked(vec![create_shell_command_sse_response(
+            shell_command.clone(),
+            Some(&working_directory),
+            Some(10_000),
+            "call_sleep",
+        )?])
+        .await;
+    create_config_toml(&codex_home, &server.uri())?;
+
+    let mut mcp = McpProcess::new(&codex_home).await?;
+    timeout(DEFAULT_READ_TIMEOUT, mcp.initialize()).await??;
+
+    let thread_req = mcp
+        .send_thread_start_request(ThreadStartParams {
+            model: Some("mock-model".to_string()),
+            ..Default::default()
+        })
+        .await?;
+    let thread_resp: JSONRPCResponse = timeout(
+        DEFAULT_READ_TIMEOUT,
+        mcp.read_stream_until_response_message(RequestId::Integer(thread_req)),
+    )
+    .await??;
+    let ThreadStartResponse { thread, .. } = to_response::<ThreadStartResponse>(thread_resp)?;
+
+    let turn_req = mcp
+        .send_turn_start_request(TurnStartParams {
+            thread_id: thread.id.clone(),
+            input: vec![V2UserInput::Text {
+                text: "run sleep".to_string(),
+                text_elements: Vec::new(),
+            }],
+            cwd: Some(working_directory.clone()),
+            ..Default::default()
+        })
+        .await?;
+    let turn_resp: JSONRPCResponse = timeout(
+        DEFAULT_READ_TIMEOUT,
+        mcp.read_stream_until_response_message(RequestId::Integer(turn_req)),
+    )
+    .await??;
+    let TurnStartResponse { turn } = to_response::<TurnStartResponse>(turn_resp)?;
+
+    let _task_started: JSONRPCNotification = timeout(
+        DEFAULT_READ_TIMEOUT,
+        mcp.read_stream_until_notification_message("codex/event/task_started"),
+    )
+    .await??;
+
+    let oversized_input = "x".repeat(MAX_USER_INPUT_TEXT_CHARS + 1);
+    let steer_req = mcp
+        .send_turn_steer_request(TurnSteerParams {
+            thread_id: thread.id.clone(),
+            input: vec![V2UserInput::Text {
+                text: oversized_input.clone(),
+                text_elements: Vec::new(),
+            }],
+            expected_turn_id: turn.id.clone(),
+        })
+        .await?;
+    let steer_err: JSONRPCError = timeout(
+        DEFAULT_READ_TIMEOUT,
+        mcp.read_stream_until_error_message(RequestId::Integer(steer_req)),
+    )
+    .await??;
+
+    assert_eq!(steer_err.error.code, INVALID_PARAMS_ERROR_CODE);
+    assert_eq!(
+        steer_err.error.message,
+        format!("Input exceeds the maximum length of {MAX_USER_INPUT_TEXT_CHARS} characters.")
+    );
+    let data = steer_err
+        .error
+        .data
+        .expect("expected structured error data");
+    assert_eq!(data["input_error_code"], INPUT_TOO_LARGE_ERROR_CODE);
+    assert_eq!(data["max_chars"], MAX_USER_INPUT_TEXT_CHARS);
+    assert_eq!(data["actual_chars"], oversized_input.chars().count());
+
+    mcp.interrupt_turn_and_wait_for_aborted(thread.id, turn.id, DEFAULT_READ_TIMEOUT)
+        .await?;
+
+    Ok(())
+}
+
 #[tokio::test]
 async fn turn_steer_returns_active_turn_id() -> Result<()> {
     #[cfg(target_os = "windows")]