142 Commits

• Generate more typescript types and return conversation id with ConversationSummary (#3219)

  This PR does multiple things that are necessary for conversation resume
  to work from the extension. I wanted to make sure everything worked so
  these changes wound up in one PR:
  1. Generate more ts types
  2. Resume rollout history files rather than create a new one every time
  it is resumed so you don't see a duplicate conversation in history for
  every resume. Chatted with @aibrahim-oai to verify this
  3. Return conversation_id in conversation summaries
  4. [Cleanup] Use serde and strong types for a lot of the rollout file
  parsing

  Gabriel Peal · 2025-09-08 17:54:47 -04:00
  5eaaf307e1
• Use ConversationId instead of raw Uuids (#3282)

  We're trying to migrate from `session_id: Uuid` to `conversation_id:
  ConversationId`. Not only does this give us more type safety but it
  unifies our terminology across Codex and with the implementation of
  session resuming, a conversation (which can span multiple sessions) is
  more appropriate.
  
  I started this impl on https://github.com/openai/codex/pull/3219 as part
  of getting resume working in the extension but it's big enough that it
  should be broken out.

  Gabriel Peal · 2025-09-07 23:22:25 -04:00
  c8fab51372
• Move token usage/context information to session level (#3221)

  Move context information into the main loop so it can be used to
  interrupt the loop or start auto-compaction.

  pakrym-oai · 2025-09-06 15:19:23 +00:00
  0269096229
• Never store requests (#3212)

  When item ids are sent to Responses API it will load them from the
  database ignoring the provided values. This adds extra latency.
  
  Not having the mode to store requests also allows us to simplify the
  code.
  
  ## Breaking change
  
  The `disable_response_storage` configuration option is removed.

  pakrym-oai · 2025-09-05 10:41:47 -07:00
  5775174ec2
• fix: add callback to map before sending request to fix race condition (#3146)

  Last week, I thought I found the smoking gun in our flaky integration
  tests where holding these locks could have led to potential deadlock:
  
  - https://github.com/openai/codex/pull/2876
  - https://github.com/openai/codex/pull/2878
  
  Yet even after those PRs went in, we continued to see flakinees in our
  integration tests! Though with the additional logging added as part of
  debugging those tests, I now saw things like:
  
  ```
  read message from stdout: Notification(JSONRPCNotification { jsonrpc: "2.0", method: "codex/event/exec_approval_request", params: Some(Object {"id": String("0"), "msg": Object {"type": String("exec_approval_request"), "call_id": String("call1"), "command": Array [String("python3"), String("-c"), String("print(42)")], "cwd": String("/tmp/.tmpFj2zwi/workdir")}, "conversationId": String("c67b32c5-9475-41bf-8680-f4b4834ebcc6")}) })
  notification: Notification(JSONRPCNotification { jsonrpc: "2.0", method: "codex/event/exec_approval_request", params: Some(Object {"id": String("0"), "msg": Object {"type": String("exec_approval_request"), "call_id": String("call1"), "command": Array [String("python3"), String("-c"), String("print(42)")], "cwd": String("/tmp/.tmpFj2zwi/workdir")}, "conversationId": String("c67b32c5-9475-41bf-8680-f4b4834ebcc6")}) })
  read message from stdout: Request(JSONRPCRequest { id: Integer(0), jsonrpc: "2.0", method: "execCommandApproval", params: Some(Object {"conversation_id": String("c67b32c5-9475-41bf-8680-f4b4834ebcc6"), "call_id": String("call1"), "command": Array [String("python3"), String("-c"), String("print(42)")], "cwd": String("/tmp/.tmpFj2zwi/workdir")}) })
  writing message to stdin: Response(JSONRPCResponse { id: Integer(0), jsonrpc: "2.0", result: Object {"decision": String("approved")} })
  in read_stream_until_notification_message(codex/event/task_complete)
  [mcp stderr] 2025-09-04T00:00:59.738585Z  INFO codex_mcp_server::message_processor: <- response: JSONRPCResponse { id: Integer(0), jsonrpc: "2.0", result: Object {"decision": String("approved")} }
  [mcp stderr] 2025-09-04T00:00:59.738740Z DEBUG codex_core::codex: Submission sub=Submission { id: "1", op: ExecApproval { id: "0", decision: Approved } }
  [mcp stderr] 2025-09-04T00:00:59.738832Z  WARN codex_core::codex: No pending approval found for sub_id: 0
  ```
  
  That is, a response was sent for a request, but no callback was in place
  to handle the response!
  
  This time, I think I may have found the underlying issue (though the
  fixes for holding locks for too long may have also been part of it),
  which is I found cases where we were sending the request:
  
  
  https://github.com/openai/codex/blob/234c0a0469db222f05df08d00ae5032312f77427/codex-rs/core/src/codex.rs#L597
  
  before inserting the `Sender` into the `pending_approvals` map (which
  has to wait on acquiring a mutex):
  
  
  https://github.com/openai/codex/blob/234c0a0469db222f05df08d00ae5032312f77427/codex-rs/core/src/codex.rs#L598-L601
  
  so it is possible the request could go out and the client could respond
  before `pending_approvals` was updated!
  
  Note this was happening in both `request_command_approval()` and
  `request_patch_approval()`, which maps to the sorts of errors we have
  been seeing when these integration tests have been flaking on us.
  
  While here, I am also adding some extra logging that prints if inserting
  into `pending_approvals` overwrites an entry as opposed to purely
  inserting one. Today, a conversation can have only one pending request
  at a time, but as we are planning to support parallel tool calls, this
  invariant may not continue to hold, in which case we need to revisit
  this abstraction.

  Michael Bolin · 2025-09-04 07:38:28 -07:00
  bd4fa85507
• Dividing UserMsgs into categories to send it back to the tui (#3127)

  This PR does the following:
  
  - divides user msgs into 3 categories: plain, user instructions, and
  environment context
  - Centralizes adding user instructions and environment context to a
  degree
  - Improve the integration testing
  
  Building on top of #3123
  
  Specifically this
  [comment](https://github.com/openai/codex/pull/3123#discussion_r2319885089).
  We need to send the user message while ignoring the User Instructions
  and Environment Context we attach.

  Ahmed Ibrahim · 2025-09-04 05:34:50 +00:00
  2b96f9f569
• Replay EventMsgs from Response Items when resuming a session with history. (#3123)

  ### Overview
  
  This PR introduces the following changes:
  	1.	Adds a unified mechanism to convert ResponseItem into EventMsg.
  2. Ensures that when a session is initialized with initial history, a
  vector of EventMsg is sent along with the session configuration. This
  allows clients to re-render the UI accordingly.
  	3. 	Added integration testing
  
  ### Caveats
  
  This implementation does not send every EventMsg that was previously
  dispatched to clients. The excluded events fall into two categories:
  	•	“Arguably” rolled-out events
  Examples include tool calls and apply-patch calls. While these events
  are conceptually rolled out, we currently only roll out ResponseItems.
  These events are already being handled elsewhere and transformed into
  EventMsg before being sent.
  	•	Non-rolled-out events
  Certain events such as TurnDiff, Error, and TokenCount are not rolled
  out at all.
  
  ### Future Directions
  
  At present, resuming a session involves maintaining two states:
  	•	UI State
  Clients can replay most of the important UI from the provided EventMsg
  history.
  	•	Model State
  The model receives the complete session history to reconstruct its
  internal state.
  
  This design provides a solid foundation. If, in the future, more precise
  UI reconstruction is needed, we have two potential paths:
  1. Introduce a third data structure that allows us to derive both
  ResponseItems and EventMsgs.
  2. Clearly divide responsibilities: the core system ensures the
  integrity of the model state, while clients are responsible for
  reconstructing the UI.

  Ahmed Ibrahim · 2025-09-04 04:47:00 +00:00
  f2036572b6
• Move CodexAuth and AuthManager to the core crate (#3074)

  Fix a long standing layering issue.

  pakrym-oai · 2025-09-02 18:36:19 -07:00
  03e2796ca4
• chore: unify history loading (#2736)

  We have two ways of loading conversation with a previous history. Fork
  conversation and the experimental resume that we had before. In this PR,
  I am unifying their code path. The path is getting the history items and
  recording them in a brand new conversation. This PR also constraint the
  rollout recorder responsibilities to be only recording to the disk and
  loading from the disk.
  
  The PR also fixes a current bug when we have two forking in a row:
  History 1:
  <Environment Context>
  UserMessage_1
  UserMessage_2
  UserMessage_3
  
  **Fork with n = 1 (only remove one element)**
  History 2:
  <Environment Context>
  UserMessage_1
  UserMessage_2
  <Environment Context>
  
  **Fork with n = 1 (only remove one element)**
  History 2:
  <Environment Context>
  UserMessage_1
  UserMessage_2
  **<Environment Context>**
  
  This shouldn't happen but because we were appending the `<Environment
  Context>` after each spawning and it's considered as _user message_.
  Now, we don't add this message if restoring and old conversation.

  Ahmed Ibrahim · 2025-09-02 22:44:29 +00:00
  431a10fc50
• fix: drop Mutex before calling tx_approve.send() (#2876)
  Michael Bolin · 2025-08-28 22:49:29 -07:00
  c988ce28fe
• Following up on #2371 post commit feedback (#2852)

  - Introduce websearch end to complement the begin 
  - Moves the logic of adding the sebsearch tool to
  create_tools_json_for_responses_api
  - Making it the client responsibility to toggle the tool on or off 
  - Other misc in #2371 post commit feedback
  - Show the query:
  
  <img width="1392" height="151" alt="image"
  src="https://github.com/user-attachments/assets/8457f1a6-f851-44cf-bcca-0d4fe460ce89"
  />

  Ahmed Ibrahim · 2025-08-28 19:24:38 -07:00
  9dbe7284d2
• Custom /prompts (#2696)

  Adds custom `/prompts` to `~/.codex/prompts/<command>.md`.
  
  <img width="239" height="107" alt="Screenshot 2025-08-25 at 6 22 42 PM"
  src="https://github.com/user-attachments/assets/fe6ebbaa-1bf6-49d3-95f9-fdc53b752679"
  />
  
  ---
  
  Details:
  
  1. Adds `Op::ListCustomPrompts` to core.
  2. Returns `ListCustomPromptsResponse` with list of `CustomPrompt`
  (name, content).
  3. TUI calls the operation on load, and populates the custom prompts
  (excluding prompts that collide with builtins).
  4. Selecting the custom prompt automatically sends the prompt to the
  agent.

  dedrisian-oai · 2025-08-29 02:16:39 +00:00
  b8e8454b3f
• Race condition in compact (#2746)

  This fixes the flakiness in
  `summarize_context_three_requests_and_instructions` because we should
  trim history before sending task complete.

  Ahmed Ibrahim · 2025-08-28 12:53:00 -07:00
  ed06f90fb3
• Add "View Image" tool (#2723)

  Adds a "View Image" tool so Codex can find and see images by itself:
  
  <img width="1772" height="420" alt="Screenshot 2025-08-26 at 10 40
  04 AM"
  src="https://github.com/user-attachments/assets/7a459c7b-0b86-4125-82d9-05fbb35ade03"
  />

  dedrisian-oai · 2025-08-27 17:41:23 -07:00
  4e9ad23864
• Bug fix: deduplicate assistant messages (#2758)

  We are treating assistant messages in a different way than other
  messages which resulted in a duplicated history.
  
  See #2698

  Ahmed Ibrahim · 2025-08-27 01:29:16 -07:00
  2d2f66f9c5
• send context window with task started (#2752)

  - Send context window with task started
  - Accounting for changing the model per turn

  Ahmed Ibrahim · 2025-08-27 00:04:21 -07:00
  d0e06f74e2
• Don't send Exec deltas on apply patch (#2742)

  We are now sending exec deltas on apply patch which doesn't make sense.

  Ahmed Ibrahim · 2025-08-26 19:16:51 -07:00
  3eb11c10d0
• Add auth to send_user_turn (#2688)

  It is there for send_user_message but was omitted from send_user_turn.
  Presumably this was a mistake

  Gabriel Peal · 2025-08-25 18:57:20 -04:00
  cb32f9c64e
• avoid error when /compact response has no token_usage (#2417) (#2640)

  **Context**  
  When running `/compact`, `drain_to_completed` would throw an error if
  `token_usage` was `None` in `ResponseEvent::Completed`. This made the
  command fail even though everything else had succeeded.
  
  **What changed**  
  - Instead of erroring, we now just check `if let Some(token_usage)`
  before sending the event.
  - If it’s missing, we skip it and move on.  
  
  **Why**  
  This makes `AgentTask::compact()` behave in the same way as
  `AgentTask::spawn()`, which also doesn’t error out when `token_usage`
  isn’t available. Keeps things consistent and avoids unnecessary
  failures.
  
  **Fixes**  
  Closes #2417
  
  ---------
  
  Co-authored-by: Ahmed Ibrahim <aibrahim@openai.com>

  Odysseas Yiakoumis · 2025-08-25 18:42:22 +00:00
  a6c346b9e1
• fix: Scope ExecSessionManager to Session instead of using global singleton (#2664)

  The `SessionManager` in `exec_command` owns a number of
  `ExecCommandSession` objects where `ExecCommandSession` has a
  non-trivial implementation of `Drop`, so we want to be able to drop an
  individual `SessionManager` to help ensure things get cleaned up in a
  timely fashion. To that end, we should have one `SessionManager` per
  session rather than one global one for the lifetime of the CLI process.

  Michael Bolin · 2025-08-24 22:52:49 -07:00
  295ca27e98
• fix: build is broken on main; introduce ToolsConfigParams to help fix (#2663)

  `ToolsConfig::new()` taking a large number of boolean params was hard to
  manage and it finally bit us (see
  https://github.com/openai/codex/pull/2660). This changes
  `ToolsConfig::new()` so that it takes a struct (and also reduces the
  visibility of some members, where possible).

  Michael Bolin · 2025-08-24 22:43:42 -07:00
  7b20db942a
• Add web search tool (#2371)

  Adds web_search tool, enabling the model to use Responses API web_search
  tool.
  - Disabled by default, enabled by --search flag
  - When --search is passed, exposes web_search_request function tool to
  the model, which triggers user approval. When approved, the model can
  use the web_search tool for the remainder of the turn
  <img width="1033" height="294" alt="image"
  src="https://github.com/user-attachments/assets/62ac6563-b946-465c-ba5d-9325af28b28f"
  />
  
  ---------
  
  Co-authored-by: easong-openai <easong@openai.com>

  Reuben Narad · 2025-08-23 22:58:56 -07:00
  363636f5eb
• send-aggregated output (#2364)

  We want to send an aggregated output of stderr and stdout so we don't
  have to aggregate it stderr+stdout as we lose order sometimes.
  
  ---------
  
  Co-authored-by: Gabriel Peal <gpeal@users.noreply.github.com>

  Ahmed Ibrahim · 2025-08-23 16:54:31 +00:00
  957d44918d
• feat: StreamableShell with exec_command and write_stdin tools (#2574)
  Michael Bolin · 2025-08-22 18:10:55 -07:00
  e3b03eaccb
• fork conversation from a previous message (#2575)

  This can be the underlying logic in order to start a conversation from a
  previous message. will need some love in the UI.
  
  Base for building this: #2588

  Ahmed Ibrahim · 2025-08-22 17:06:09 -07:00
  311ad0ce26
• tui: coalesce command output; show unabridged commands in transcript (#2590)

  https://github.com/user-attachments/assets/effec7c7-732a-4b61-a2ae-3cb297b6b19b

  Jeremy Rose · 2025-08-22 16:32:31 -07:00
  d994019f3f
• Move models.rs to protocol (#2595)

  Moving models.rs to protocol so we can use them in `Codex` operations

  Ahmed Ibrahim · 2025-08-22 22:18:54 +00:00
  097782c775
• fix: prefer sending MCP structuredContent as the function call response, if available (#2594)

  Prior to this change, when we got a `CallToolResult` from an MCP server,
  we JSON-serialized its `content` field as the `content` to send back to
  the model as part of the function call output that we send back to the
  model. This meant that we were dropping the `structuredContent` on the
  floor.
  
  Though reading
  https://modelcontextprotocol.io/specification/2025-06-18/schema#tool, it
  appears that if `outputSchema` is specified, then `structuredContent`
  should be set, which seems to be a "higher-fidelity" response to the
  function call. This PR updates our handling of `CallToolResult` to
  prefer using the JSON-serialization of `structuredContent`, if present,
  using `content` as a fallback.
  
  Also, it appears that the sense of `success` was inverted prior to this
  PR!

  Michael Bolin · 2025-08-22 14:10:18 -07:00
  8ba8089592
• [apply_patch] freeform apply_patch tool (#2576)

  ## Summary
  GPT-5 introduced the concept of [custom
  tools](https://platform.openai.com/docs/guides/function-calling#custom-tools),
  which allow the model to send a raw string result back, simplifying
  json-escape issues. We are migrating gpt-5 to use this by default.
  
  However, gpt-oss models do not support custom tools, only normal
  functions. So we keep both tool definitions, and provide whichever one
  the model family supports.
  
  ## Testing
  - [x] Tested locally with various models
  - [x] Unit tests pass

  Dylan · 2025-08-22 13:42:34 -07:00
  236c4f76a6
• Add AuthManager and enhance GetAuthStatus command (#2577)

  This PR adds a central `AuthManager` struct that manages the auth
  information used across conversations and the MCP server. Prior to this,
  each conversation and the MCP server got their own private snapshots of
  the auth information, and changes to one (such as a logout or token
  refresh) were not seen by others.
  
  This is especially problematic when multiple instances of the CLI are
  run. For example, consider the case where you start CLI 1 and log in to
  ChatGPT account X and then start CLI 2 and log out and then log in to
  ChatGPT account Y. The conversation in CLI 1 is still using account X,
  but if you create a new conversation, it will suddenly (and
  unexpectedly) switch to account Y.
  
  With the `AuthManager`, auth information is read from disk at the time
  the `ConversationManager` is constructed, and it is cached in memory.
  All new conversations use this same auth information, as do any token
  refreshes.
  
  The `AuthManager` is also used by the MCP server's GetAuthStatus
  command, which now returns the auth method currently used by the MCP
  server.
  
  This PR also includes an enhancement to the GetAuthStatus command. It
  now accepts two new (optional) input parameters: `include_token` and
  `refresh_token`. Callers can use this to request the in-use auth token
  and can optionally request to refresh the token.
  
  The PR also adds tests for the login and auth APIs that I recently added
  to the MCP server.

  Eric Traut · 2025-08-22 13:10:11 -07:00
  dc42ec0eb4
• Parse and expose stream errors (#2540)
  easong-openai · 2025-08-21 01:15:24 -07:00
  8ad56be06e
• [prompt] xml-format EnvironmentContext (#2272)

  ## Summary
  Before we land #2243, let's start printing environment_context in our
  preferred format. This struct will evolve over time with new
  information, xml gives us a balance of human readable without too much
  parsing, llm readable, and extensible.
  
  Also moves us over to an Option-based struct, so we can easily provide
  diffs to the model.
  
  ## Testing
  - [x] Updated tests to reflect new format

  Dylan · 2025-08-20 23:45:16 -07:00
  d2b2a6d13a
• Bridge command generation to powershell when on Windows (#2319)

  ## What? Why? How?
  - When running on Windows, codex often tries to invoke bash commands,
  which commonly fail (unless WSL is installed)
  - Fix: Detect if powershell is available and, if so, route commands to
  it
  - Also add a shell_name property to environmental context for codex to
  default to powershell commands when running in that environment
  
  ## Testing
  - Tested within WSL and powershell (e.g. get top 5 largest files within
  a folder and validated that commands generated were powershell commands)
  - Tested within Zsh
  - Updated unit tests
  
  ---------
  
  Co-authored-by: Eddy Escardo <eddy@openai.com>

  eddy-win · 2025-08-20 16:30:34 -07:00
  050b9baeb6
• chore: upgrade to Rust 1.89 (#2465)

  Codex created this PR from the following prompt:
  
  > upgrade this entire repo to Rust 1.89. Note that this requires
  updating codex-rs/rust-toolchain.toml as well as the workflows in
  .github/. Make sure that things are "clippy clean" as this change will
  likely uncover new Clippy errors. `just fmt` and `cargo clippy --tests`
  are sufficient to check for correctness
  
  Note this modifies a lot of lines because it folds nested `if`
  statements using `&&`.
  
  ---
  [//]: # (BEGIN SAPLING FOOTER)
  Stack created with [Sapling](https://sapling-scm.com). Best reviewed
  with [ReviewStack](https://reviewstack.dev/openai/codex/pull/2465).
  * #2467
  * __->__ #2465

  Michael Bolin · 2025-08-19 13:22:02 -07:00
  50c48e88f5
• [tui] Support /mcp command (#2430)

  ## Summary
  Adds a `/mcp` command to list active tools. We can extend this command
  to allow configuration of MCP tools, but for now a simple list command
  will help debug if your config.toml and your tools are working as
  expected.

  Dylan · 2025-08-19 09:00:31 -07:00
  e7e5fe91c8
• Add an operation to override current task context (#2431)

  - Added an operation to override current task context
  - Added a test to check that cache stays the same

  Ahmed Ibrahim · 2025-08-18 19:59:19 +00:00
  c283f9f6ce
• consolidate reasoning enums into one (#2428)

  We have three enums for each of reasoning summaries and reasoning effort
  with same values. They can be consolidated into one.

  Ahmed Ibrahim · 2025-08-18 11:50:17 -07:00
  c9963b52e9
• fix: introduce EventMsg::TurnAborted (#2365)

  Introduces `EventMsg::TurnAborted` that should be sent in response to
  `Op::Interrupt`.
  
  In the MCP server, updates the handling of a
  `ClientRequest::InterruptConversation` request such that it sends the
  `Op::Interrupt` but does not respond to the request until it sees an
  `EventMsg::TurnAborted`.

  Michael Bolin · 2025-08-17 21:40:31 -07:00
  b581498882
• fix: introduce codex-protocol crate (#2355)
  Michael Bolin · 2025-08-15 12:44:40 -07:00
  d262244725
• feat: introduce Op:UserTurn (#2329)

  This introduces `Op::UserTurn`, which makes it possible to override many
  of the fields that were set when the `Session` was originally created
  when creating a new conversation turn. This is one way we could support
  changing things like `model` or `cwd` in the middle of the conversation,
  though we may want to consider making each field optional, or
  alternatively having a separate `Op` that mutates the `TurnContext`
  associated with a `submission_loop()`.
  
  ---
  [//]: # (BEGIN SAPLING FOOTER)
  Stack created with [Sapling](https://sapling-scm.com). Best reviewed
  with [ReviewStack](https://reviewstack.dev/openai/codex/pull/2329).
  * #2345
  * __->__ #2329
  * #2343
  * #2340
  * #2338

  Michael Bolin · 2025-08-15 09:56:05 -07:00
  17aa394ae7
• feat: introduce TurnContext (#2343)

  This PR introduces `TurnContext`, which is designed to hold a set of
  fields that should be constant for a turn of a conversation. Note that
  the fields of `TurnContext` were previously governed by `Session`.
  
  Ultimately, we want to enable users to change these values between turns
  (changing model, approval policy, etc.), though in the current
  implementation, the `TurnContext` is constant for the entire
  conversation.
  
  ---
  [//]: # (BEGIN SAPLING FOOTER)
  Stack created with [Sapling](https://sapling-scm.com). Best reviewed
  with [ReviewStack](https://reviewstack.dev/openai/codex/pull/2345).
  * #2345
  * #2329
  * __->__ #2343
  * #2340
  * #2338

  Michael Bolin · 2025-08-15 09:40:02 -07:00
  13ed67cfc1
• fix: introduce MutexExt::lock_unchecked() so we stop ignoring unwrap() throughout codex.rs (#2340)

  This way we are sure a dangerous `unwrap()` does not sneak in!
  
  ---
  [//]: # (BEGIN SAPLING FOOTER)
  Stack created with [Sapling](https://sapling-scm.com). Best reviewed
  with [ReviewStack](https://reviewstack.dev/openai/codex/pull/2340).
  * #2345
  * #2329
  * #2343
  * __->__ #2340
  * #2338

  Michael Bolin · 2025-08-15 09:14:44 -07:00
  6730592433
• fix: tighten up checks against writable folders for SandboxPolicy (#2338)

  I was looking at the implementation of `Session::get_writable_roots()`,
  which did not seem right, as it was a copy of writable roots, which is
  not guaranteed to be in sync with the `sandbox_policy` field.
  
  I looked at who was calling `get_writable_roots()` and its only call
  site was `apply_patch()` in `codex-rs/core/src/apply_patch.rs`, which
  took the roots and forwarded them to `assess_patch_safety()` in
  `safety.rs`. I updated `assess_patch_safety()` to take `sandbox_policy:
  &SandboxPolicy` instead of `writable_roots: &[PathBuf]` (and replaced
  `Session::get_writable_roots()` with `Session::get_sandbox_policy()`).
  
  Within `safety.rs`, it was fairly easy to update
  `is_write_patch_constrained_to_writable_paths()` to work with
  `SandboxPolicy`, and in particular, it is far more accurate because, for
  better or worse, `SandboxPolicy::get_writable_roots_with_cwd()` _returns
  an empty vec_ for `SandboxPolicy::DangerFullAccess`, suggesting that
  _nothing_ is writable when in reality _everything_ is writable. With
  this PR, `is_write_patch_constrained_to_writable_paths()` now does the
  right thing for each variant of `SandboxPolicy`.
  
  I thought this would be the end of the story, but it turned out that
  `test_writable_roots_constraint()` in `safety.rs` needed to be updated,
  as well. In particular, the test was writing to
  `std::env::current_dir()` instead of a `TempDir`, which I suspect was a
  holdover from earlier when `SandboxPolicy::WorkspaceWrite` would always
  make `TMPDIR` writable on macOS, which made it hard to write tests to
  verify `SandboxPolicy` in `TMPDIR`. Fortunately, we now have
  `exclude_tmpdir_env_var` as an option on
  `SandboxPolicy::WorkspaceWrite`, so I was able to update the test to
  preserve the existing behavior, but to no longer write to
  `std::env::current_dir()`.
  
  
  
  
  
  
  
  ---
  [//]: # (BEGIN SAPLING FOOTER)
  Stack created with [Sapling](https://sapling-scm.com). Best reviewed
  with [ReviewStack](https://reviewstack.dev/openai/codex/pull/2338).
  * #2345
  * #2329
  * #2343
  * #2340
  * __->__ #2338

  Michael Bolin · 2025-08-15 09:06:15 -07:00
  26c8373821
• [tools] Add apply_patch tool (#2303)

  ## Summary
  We've been seeing a number of issues and reports with our synthetic
  `apply_patch` tool, e.g. #802. Let's make this a real tool - in my
  anecdotal testing, it's critical for GPT-OSS models, but I'd like to
  make it the standard across GPT-5 and codex models as well.
  
  ## Testing
  - [x] Tested locally
  - [x] Integration test

  Dylan · 2025-08-15 11:55:53 -04:00
  6df8e35314
• Added allow-expect-in-tests / allow-unwrap-in-tests (#2328)

  This PR:
  * Added the clippy.toml to configure allowable expect / unwrap usage in
  tests
  * Removed as many expect/allow lines as possible from tests
  * moved a bunch of allows to expects where possible
  
  Note: in integration tests, non `#[test]` helper functions are not
  covered by this so we had to leave a few lingering `expect(expect_used`
  checks around

  Parker Thompson · 2025-08-14 17:59:01 -07:00
  a075424437
• fix: parallelize logic in Session::new() (#2305)

  #2291 made it so that `Session::new()` is on the critical path to
  `Codex::spawn()`, which means it is on the hot path to CLI startup. This
  refactors `Session::new()` to run a number of async tasks in parallel
  that were previously run serially to try to reduce latency.

  Michael Bolin · 2025-08-14 13:29:58 -07:00
  8f11652458
• [context] Store context messages in rollouts (#2243)

  ## Summary
  Currently, we use request-time logic to determine the user_instructions
  and environment_context messages. This means that neither of these
  values can change over time as conversations go on. We want to add in
  additional details here, so we're migrating these to save these messages
  to the rollout file instead. This is simpler for the client, and allows
  us to append additional environment_context messages to each turn if we
  want
  
  ## Testing
  - [x] Integration test coverage
  - [x] Tested locally with a few turns, confirmed model could reference
  environment context and cached token metrics were reasonably high

  Dylan · 2025-08-14 14:51:13 -04:00
  544980c008
• exploration: create Session as part of Codex::spawn() (#2291)

  Historically, `Codex::spawn()` would create the instance of `Codex` and
  enforce, by construction, that `Op::ConfigureSession` was the first `Op`
  submitted via `submit()`. Then over in `submission_loop()`, it would
  handle the case for taking the parameters of `Op::ConfigureSession` and
  turning it into a `Session`.
  
  This approach has two challenges from a state management perspective:
  
  
  https://github.com/openai/codex/blob/f968a1327ad39a7786759ea8f1d1c088fe41e91b/codex-rs/core/src/codex.rs#L718
  
  - The local `sess` variable in `submission_loop()` has to be `mut` and
  `Option<Arc<Session>>` because it is not invariant that a `Session` is
  present for the lifetime of the loop, so there is a lot of logic to deal
  with the case where `sess` is `None` (e.g., the `send_no_session_event`
  function and all of its callsites).
  - `submission_loop()` is written in such a way that
  `Op::ConfigureSession` could be observed multiple times, but in
  practice, it is only observed exactly once at the start of the loop.
  
  In this PR, we try to simplify the state management by _removing_ the
  `Op::ConfigureSession` enum variant and constructing the `Session` as
  part of `Codex::spawn()` so that it can be passed to `submission_loop()`
  as `Arc<Session>`. The original logic from the `Op::ConfigureSession`
  has largely been moved to the new `Session::new()` constructor.
  
  ---
  
  Incidentally, I also noticed that the handling of `Op::ConfigureSession`
  can result in events being dispatched in addition to
  `EventMsg::SessionConfigured`, as an `EventMsg::Error` is created for
  every MCP initialization error, so it is important to preserve that
  behavior:
  
  
  https://github.com/openai/codex/blob/f968a1327ad39a7786759ea8f1d1c088fe41e91b/codex-rs/core/src/codex.rs#L901-L916
  
  Though admittedly, I believe this does not play nice with #2264, as
  these error messages will likely be dispatched before the client has a
  chance to call `addConversationListener`, so we likely need to make it
  so `newConversation` automatically creates the subscription, but we must
  also guarantee that the "ack" from `newConversation` is returned before
  any other conversation-related notifications are sent so the client
  knows what `conversation_id` to match on.

  Michael Bolin · 2025-08-14 09:55:28 -07:00
  cf7a7e63a3
• fix: make all fields of Session private (#2285)

  As `Session` needs a bit of work, it will make things easier to move
  around if we can start by reducing the extent of its public API. This
  makes all the fields private, though adds three `pub(crate)` getters.
  
  ---
  [//]: # (BEGIN SAPLING FOOTER)
  Stack created with [Sapling](https://sapling-scm.com). Best reviewed
  with [ReviewStack](https://reviewstack.dev/openai/codex/pull/2285).
  * #2287
  * #2286
  * __->__ #2285

  Michael Bolin · 2025-08-13 22:53:54 -07:00
  085f166707
• Parse reasoning text content (#2277)

  Sometimes COT is returns as text content instead of `ReasoningText`. We
  should parse it but not serialize back on requests.
  
  ---------
  
  Co-authored-by: Ahmed Ibrahim <aibrahim@openai.com>

  pakrym-oai · 2025-08-13 18:39:58 -07:00
  f1be7978cf