followup to https://github.com/openai/codex/pull/13212 to expose fast
tier controls to app server
(majority of this PR is generated schema jsons - actual code is +69 /
-35 and +24 tests )
- add service tier fields to the app-server protocol surfaces used by
thread lifecycle, turn start, config, and session configured events
- thread service tier through the app-server message processor and core
thread config snapshots
- allow runtime config overrides to carry service tier for app-server
callers
cleanup:
- Removing useless "legacy" code supporting "standard" - we moved to
None | "fast", so "standard" is not needed.
- add a local Fast mode setting in codex-core (similar to how model id
is currently stored on disk locally)
- send `service_tier=priority` on requests when Fast is enabled
- add `/fast` in the TUI and persist it locally
- feature flag
## Summary
This change removes the compiled permissions field from skill metadata
and keeps permission_profile as the single source of truth.
Skill loading no longer compiles skill permissions eagerly. Instead, the
zsh-fork skill escalation path compiles `skill.permission_profile` when
it needs to determine the sandbox to apply for a skill script.
## Behavior change
For skills that declare:
```
permissions: {}
```
we now treat that the same as having no skill permissions override,
instead of creating and using a default readonly sandbox. This change
makes the behavior more intuitive:
- only non-empty skill permission profiles affect sandboxing
- omitting permissions and writing permissions: {} now mean the same
thing
- skill metadata keeps a single permissions representation instead of
storing derived state too
Overall, this makes skill sandbox behavior easier to understand and more
predictable.
- migrate the realtime websocket transport to the new session and
handoff flow
- make the realtime model configurable in config.toml and use API-key
auth for the websocket
---------
Co-authored-by: Codex <noreply@openai.com>
Currently `thread/name/set` does only work for loaded threads.
Expand the scope to also support persisted but not-yet-loaded ones for a
more predictable API surface.
This will make it possible to rename threads discovered via
`thread/list` and similar operations.
## Summary
Codex discovered this one - shell_snapshot tests were breaking on my
machine because I had a multiline env var. We should handle these!
## Testing
- [x] existing tests pass
- [x] Updated unit tests
Fixes#13076
This PR fixes a bug that causes command-line config overrides for MCP
subtables to not be merged correctly.
Summary
- make project trust loading go through the dedicated struct so CLI
overrides can update trusted project-local MCP transports
---------
Co-authored-by: jif-oai <jif@openai.com>
## Summary
- reuse the parent shell snapshot when spawning/forking/resuming
`SessionSource::SubAgent(SubAgentSource::ThreadSpawn { .. })` sessions
- plumb inherited snapshot through `AgentControl -> ThreadManager ->
Codex::spawn -> SessionConfiguration`
- skip shell snapshot refresh on cwd updates for thread-spawn subagents
so inherited snapshots are not replaced
## Why
- avoids per-subagent shell snapshot creation and cleanup work
- keeps thread-spawn subagents on the parent snapshot path, matching the
intended parent/child snapshot model
## Validation
- `just fmt` (in `codex-rs`)
- `cargo test -p codex-core --no-run`
- `cargo test -p codex-core spawn_agent -- --nocapture`
- `cargo test -p codex-core --test all
suite::agent_jobs::spawn_agents_on_csv_runs_and_exports`
## Notes
- full `cargo test -p codex-core --test all` was left running separately
for broader verification
Co-authored-by: Codex <noreply@openai.com>
## Summary
- record a realtime close developer message when a new realtime session
replaces an active one
- assert the replacement marker through the mocked responses request
path
---------
Co-authored-by: Codex <noreply@openai.com>
Co-authored-by: Charles Cunningham <ccunningham@openai.com>
…with launchservicesd
Add mach lookup for `launchservicesd` when extending the sandbox for
`MacOSAutomationPermission::BundleIDs`. This is necessary so that the
target application can be launched for automation.
This omission was due to a spec error in a document, which has been
fixed.
Support loading plugins.
Plugins can now be enabled via [plugins.<name>] in config.toml. They are
loaded as first-class entities through PluginsManager, and their default
skills/ and .mcp.json contributions are integrated into the existing
skills and MCP flows.
## Why
[#12964](https://github.com/openai/codex/pull/12964) added
`host_executable()` support to `codex-execpolicy`, and
[#13046](https://github.com/openai/codex/pull/13046) adopted it in the
zsh-fork interception path.
The remaining gap was the preflight execpolicy check in
`core/src/exec_policy.rs`. That path derives approval requirements
before execution for `shell`, `shell_command`, and `unified_exec`, but
it was still using the default exact-token matcher.
As a result, a command that already included an absolute executable
path, such as `/usr/bin/git status`, could still miss a basename rule
like `prefix_rule(pattern = ["git"], ...)` during preflight even when
the policy also defined a matching `host_executable(name = "git", ...)`
entry.
This PR brings the same opt-in `host_executable()` resolution to the
preflight approval path when an absolute program path is already present
in the parsed command.
## What Changed
- updated
`ExecPolicyManager::create_exec_approval_requirement_for_command()` in
`core/src/exec_policy.rs` to use `check_multiple_with_options(...)` with
`MatchOptions { resolve_host_executables: true }`
- kept the existing shell parsing flow for approval derivation, but now
allow basename rules to match absolute executable paths during preflight
when `host_executable()` permits it
- updated requested-prefix amendment evaluation to use the same
host-executable-aware matching mode, so suggested `prefix_rule()`
amendments are checked consistently for absolute-path commands
- added preflight coverage for:
- absolute-path commands that should match basename rules through
`host_executable()`
- absolute-path commands whose paths are not in the allowed
`host_executable()` mapping
- requested prefix-rule amendments for absolute-path commands
## Verification
- `just fix -p codex-core`
- `cargo test -p codex-core --lib exec_policy::tests::`
## Summary
- skip online model refresh for subagent sessions
- avoid rollout flushes during subagent startup
- keep /models refresh for non-subagent sessions
## Testing
- cargo test -p codex-core --test all
suite::models_etag_responses::refresh_models_on_models_etag_mismatch_and_avoid_duplicate_models_fetch
- cargo test -p codex-core --test all
suite::remote_models::remote_models_long_model_slug_is_sent_with_high_reasoning
- cargo test -p codex-core --test all
suite::model_switching::model_switch_to_smaller_model_updates_token_context_window
- cargo test -p codex-core --test all
suite::compact::pre_sampling_compact_runs_on_switch_to_smaller_context_model
- cargo test -p codex-core --test all
suite::compact::pre_sampling_compact_runs_after_resume_and_switch_to_smaller_model
- cargo test -p codex-core --test all
suite::personality::remote_model_friendly_personality_instructions_with_feature
---------
Co-authored-by: Codex <noreply@openai.com>
## Why
- tighten Codex memory-read behavior around stale facts and conflicting
memory
- encode the risk-of-drift vs verification-effort decision rule directly
in the read-path prompt
- make partial stale-detail updates explicit so correcting only the
answer is not treated as sufficient
## What changed
- update `codex-rs/core/templates/memories/read_path.md`
- add guidance for when to verify cheap local facts vs when to answer
from older memory with visible provenance
- strengthen same-turn `MEMORY.md` updates when stored concrete details
are stale
## Notes
- this is based on some staleness eval work
## Why
[#12964](https://github.com/openai/codex/pull/12964) added
`host_executable()` support to `codex-execpolicy`, but the zsh-fork
interception path in `unix_escalation.rs` was still evaluating commands
with the default exact-token matcher.
That meant an intercepted absolute executable such as `/usr/bin/git
status` could still miss basename rules like `prefix_rule(pattern =
["git", "status"])`, even when the policy also defined a matching
`host_executable(name = "git", ...)` entry.
This PR adopts the new matching behavior in the zsh-fork runtime only.
That keeps the rollout intentionally narrow: zsh-fork already requires
explicit user opt-in, so it is a safer first caller to exercise the new
`host_executable()` scheme before expanding it to other execpolicy call
sites.
It also brings zsh-fork back in line with the current `prefix_rule()`
execution model. Until prefix rules can carry their own permission
profiles, a matched `prefix_rule()` is expected to rerun the intercepted
command unsandboxed on `allow`, or after the user accepts `prompt`,
instead of merely continuing inside the inherited shell sandbox.
## What Changed
- added `evaluate_intercepted_exec_policy()` in
`core/src/tools/runtimes/shell/unix_escalation.rs` to centralize
execpolicy evaluation for intercepted commands
- switched intercepted direct execs in the zsh-fork path to
`check_multiple_with_options(...)` with `MatchOptions {
resolve_host_executables: true }`
- added `commands_for_intercepted_exec_policy()` so zsh-fork policy
evaluation works from intercepted `(program, argv)` data instead of
reconstructing a synthetic command before matching
- left shell-wrapper parsing intentionally disabled by default behind
`ENABLE_INTERCEPTED_EXEC_POLICY_SHELL_WRAPPER_PARSING`, so
path-sensitive matching relies on later direct exec interception rather
than shell-script parsing
- made matched `prefix_rule()` decisions rerun intercepted commands with
`EscalationExecution::Unsandboxed`, while unmatched-command fallback
keeps the existing sandbox-preserving behavior
- extracted the zsh-fork test harness into
`core/tests/common/zsh_fork.rs` so both the skill-focused and
approval-focused integration suites can exercise the same runtime setup
- limited this change to the intercepted zsh-fork path rather than
changing every execpolicy caller at once
- added runtime coverage in
`core/src/tools/runtimes/shell/unix_escalation_tests.rs` for allowed and
disallowed `host_executable()` mappings and the wrapper-parsing modes
- added integration coverage in `core/tests/suite/approvals.rs` to
verify a saved `prefix_rule(pattern=["touch"], decision="allow")` reruns
under zsh-fork outside a restrictive `WorkspaceWrite` sandbox
---
[//]: # (BEGIN SAPLING FOOTER)
Stack created with [Sapling](https://sapling-scm.com). Best reviewed
with [ReviewStack](https://reviewstack.dev/openai/codex/pull/13046).
* #13065
* __->__ #13046
- override startup tooltips with model availability NUX and persist
per-model show counts in config
- stop showing each model after four exposures and fall back to normal
tooltips
Addresses https://github.com/openai/codex/issues/13007 and
https://github.com/openai/codex/issues/12170
There are situations where the ChatGPT auth backend might return a JWT
that contains no plan information. Most code paths already handle this
case well, but the internal implementation of the "account/read" app
server call was failing in this case (returning an error rather than
properly returning None for the plan).
This resulted in a situation where users needed to log in every time the
extension or app started even if they successfully logged in the last
time.
Summary
- allow ChatGPT-authenticated accounts to fall back to
`AccountPlanType::Unknown` when the token omits the plan claim
- add regression coverage in `app-server/tests/suite/v2/account.rs` to
confirm `account/read` returns `plan_type: Unknown` when the claim is
absent
- ensure the Rust auth helpers and fixtures treat missing plan claims as
Optional and default to `Unknown`
## Summary
This PR unifies rollout history reconstruction and resume/fork metadata
hydration under a single `Session::reconstruct_history_from_rollout`
implementation.
The key change from main is that replay metadata now comes from the same
reconstruction pass that rebuilds model-visible history, instead of
doing a second bespoke rollout scan to recover `previous_model` /
`reference_context_item`.
## What Changed
### Unified reconstruction output
`reconstruct_history_from_rollout` now returns a single
`RolloutReconstruction` bundle containing:
- rebuilt `history`
- `previous_model`
- `reference_context_item`
Resume and fork both consume that shared output directly.
### Reverse replay core
The reconstruction logic moved into
`codex-rs/core/src/codex/rollout_reconstruction.rs` and now scans
rollout items newest-to-oldest.
That reverse pass:
- derives `previous_model`
- derives whether `reference_context_item` is preserved or cleared
- stops early once it has both resume metadata and a surviving
`replacement_history` checkpoint
History materialization is still bridged eagerly for now by replaying
only the surviving suffix forward, which keeps the history result stable
while moving the control flow toward the future lazy reverse loader
design.
### Removed bespoke context lookup
This deletes `last_rollout_regular_turn_context_lookup` and its separate
compaction-aware scan.
The previous model / baseline metadata is now computed from the same
replay state that rebuilds history, so resume/fork cannot drift from the
reconstructed transcript view.
### `TurnContextItem` persistence contract
`TurnContextItem` is now treated as the replay source of truth for
durable model-visible baselines.
This PR keeps the following contract explicit:
- persist `TurnContextItem` for the first real user turn so resume can
recover `previous_model`
- persist it for later turns that emit model-visible context updates
- if mid-turn compaction reinjects full initial context into replacement
history, persist a fresh `TurnContextItem` after `Compacted` so
resume/fork can re-establish the baseline from the rewritten history
- do not treat manual compaction or pre-sampling compaction as creating
a new durable baseline on their own
## Behavior Preserved
- rollback replay stays aligned with `drop_last_n_user_turns`
- rollback skips only user turns
- incomplete active user turns are dropped before older finalized turns
when rollback applies
- unmatched aborts do not consume the current active turn
- missing abort IDs still conservatively clear stale compaction state
- compaction clears `reference_context_item` until a later
`TurnContextItem` re-establishes it
- `previous_model` still comes from the newest surviving user turn that
established one
## Tests
Targeted validation run for the current branch shape:
- `cd codex-rs && cargo test -p codex-core --lib
codex::rollout_reconstruction_tests -- --nocapture`
- `cd codex-rs && just fmt`
The branch also extracts the rollout reconstruction tests into
`codex-rs/core/src/codex/rollout_reconstruction_tests.rs` so this logic
has a dedicated home instead of living inline in `codex.rs`.
## Why
`execpolicy` currently keys `prefix_rule()` matching off the literal
first token. That works for rules like `["/usr/bin/git"]`, but it means
shared basename rules such as `["git"]` do not help when a caller passes
an absolute executable path like `/usr/bin/git`.
This PR lays the groundwork for basename-aware matching without changing
existing callers yet. It adds typed host-executable metadata and an
opt-in resolution path in `codex-execpolicy`, so a follow-up PR can
adopt the new behavior in `unix_escalation.rs` and other call sites
without having to redesign the policy layer first.
## What Changed
- added `host_executable(name = ..., paths = [...])` to the execpolicy
parser and validated it with `AbsolutePathBuf`
- stored host executable mappings separately from prefix rules inside
`Policy`
- added `MatchOptions` and opt-in `*_with_options()` APIs that preserve
existing behavior by default
- implemented exact-first matching with optional basename fallback,
gated by `host_executable()` allowlists when present
- normalized executable names for cross-platform matching so Windows
paths like `git.exe` can satisfy `host_executable(name = "git", ...)`
- updated `match` / `not_match` example validation to exercise the
host-executable resolution path instead of only raw prefix-rule matching
- preserved source locations for deferred example-validation errors so
policy load failures still point at the right file and line
- surfaced `resolvedProgram` on `RuleMatch` so callers can tell when a
basename rule matched an absolute executable path
- preserved host executable metadata when requirements policies overlay
file-based policies in `core/src/exec_policy.rs`
- documented the new rule shape and CLI behavior in
`execpolicy/README.md`
## Verification
- `cargo test -p codex-execpolicy`
- added coverage in `execpolicy/tests/basic.rs` for parsing, precedence,
empty allowlists, basename fallback, exact-match precedence, and
host-executable-backed `match` / `not_match` examples
- added a regression test in `core/src/exec_policy.rs` to verify
requirements overlays preserve `host_executable()` metadata
- verified `cargo test -p codex-core --lib`, including source-rendering
coverage for deferred validation errors
## Summary
- Promote ANSI-16 to truecolor for diff rendering when running inside
Windows Terminal
- Respect explicit `FORCE_COLOR` override, skipping promotion when set
- Extract a pure `diff_color_level_for_terminal` function for
testability
- Strip background tints from ANSI-16 diff output, rendering add/delete
lines with foreground color only
- Introduce `RichDiffColorLevel` to type-safely restrict background
fills to truecolor and ansi256
## Problem
Windows Terminal fully supports 24-bit (truecolor) rendering but often
does not provide the usual TERM metadata (`TERM`, `TERM_PROGRAM`,
`COLORTERM`) in `cmd.exe`/PowerShell sessions. In those environments,
`supports-color` can report only ANSI-16 support. The diff renderer
therefore falls back to a 16-color palette, producing washed-out,
hard-to-read diffs.
The screenshots below demonstrate that both PowerShell and cmd.exe don't
set any `*TERM*` environment variables.
| PowerShell | cmd.exe |
|---|---|
| <img width="2032" height="1162" alt="SCR-20260226-nfvy"
src="https://github.com/user-attachments/assets/59e968cc-4add-4c7b-a415-07163297e86a"
/> | <img width="2032" height="1162" alt="SCR-20260226-nfyc"
src="https://github.com/user-attachments/assets/d06b3e39-bf91-4ce3-9705-82bf9563a01b"
/> |
## Mental model
`StdoutColorLevel` (from `supports-color`) is the _detected_ capability.
`DiffColorLevel` is the _intended_ capability for diff rendering. A new
intermediary — `diff_color_level_for_terminal` — maps one to the other
and is the single place where terminal-specific overrides live.
Windows Terminal is detected two independent ways: the `TerminalName`
parsed by `terminal_info()` and the raw presence of `WT_SESSION`. When
`WT_SESSION` is present and `FORCE_COLOR` is not set, we promote
unconditionally to truecolor. When `WT_SESSION` is absent but
`TerminalName::WindowsTerminal` is detected, we promote only the ANSI-16
level (not `Unknown`).
A single override helper — `has_force_color_override()` — checks whether
`FORCE_COLOR` is set. When it is, both the `WT_SESSION` fast-path and
the `TerminalName`-based promotion are suppressed, preserving explicit
user intent.
| PowerShell | cmd.exe | WSL | Bash for Windows |
|---|---|---|---|
|
|
|
|
|
## Non-goals
- This does not change color detection for anything outside the diff
renderer (e.g. the chat widget, markdown rendering).
- This does not add a user-facing config knob; `FORCE_COLOR` already
serves that role.
## Tradeoffs
- The `has_wt_session` signal is intentionally kept separate from
`TerminalName::WindowsTerminal`. `terminal_info()` is derived with
`TERM_PROGRAM` precedence, so it can differ from raw `WT_SESSION`.
- Real-world validation in this issue: in both `cmd.exe` and PowerShell,
`TERM`/`TERM_PROGRAM`/`COLORTERM` were absent, so TERM-based capability
hints were unavailable in those sessions.
- Checking `FORCE_COLOR` for presence rather than parsing its value is a
simplification. In practice `supports-color` has already parsed it, so
our check is a coarse "did the user set _anything_?" gate. The effective
color level still comes from `supports-color`.
- When `WT_SESSION` is present without `FORCE_COLOR`, we promote to
truecolor regardless of `stdout_level` (including `Unknown`). This is
aggressive but correct: `WT_SESSION` is a strong signal that we're in
Windows Terminal.
- ANSI-16 add/delete backgrounds (bright green/red) overpower
syntax-highlighted token colors, making diffs harder to read.
Foreground-only cues (colored text, gutter signs) preserve readability
on low-color terminals.
## Architecture
```
stdout_color_level() ──┐
terminal_info().name ──┤
WT_SESSION presence ──┼──▶ diff_color_level_for_terminal() ──▶ DiffColorLevel
FORCE_COLOR presence ──┘ │
▼
RichDiffColorLevel::from_diff_color_level()
│
┌──────────┴──────────┐
│ Some(TrueColor|256) │ → bg tints
│ None (Ansi16) │ → fg only
└─────────────────────┘
```
`diff_color_level()` is the environment-reading entry point; it gathers
the four runtime signals and delegates to the pure, testable
`diff_color_level_for_terminal()`.
## Observability
No new logs or metrics. Incorrect color selection is immediately visible
as broken diff rendering; the test suite covers the decision matrix
exhaustively.
## Tests
Six new unit tests exercise every branch of
`diff_color_level_for_terminal`:
| Test | Inputs | Expected |
|------|--------|----------|
| `windows_terminal_promotes_ansi16_to_truecolor_for_diffs` | Ansi16 +
WindowsTerminal name | TrueColor |
| `wt_session_promotes_ansi16_to_truecolor_for_diffs` | Ansi16 +
WT_SESSION only | TrueColor |
| `non_windows_terminal_keeps_ansi16_diff_palette` | Ansi16 + WezTerm |
Ansi16 |
| `wt_session_promotes_unknown_color_level_to_truecolor` | Unknown +
WT_SESSION | TrueColor |
| `explicit_force_override_keeps_ansi16_on_windows_terminal` | Ansi16 +
WindowsTerminal + FORCE_COLOR | Ansi16 |
| `explicit_force_override_keeps_ansi256_on_windows_terminal` | Ansi256
+ WT_SESSION + FORCE_COLOR | Ansi256 |
| `ansi16_add_style_uses_foreground_only` | Dark + Ansi16 | fg=Green,
bg=None |
| (and any other new snapshot/assertion tests from commits d757fee and
d7c78b3) | | |
## Test plan
- [x] Verify all new unit tests pass (`cargo test -p codex-tui --lib`)
- [x] On Windows Terminal: confirm diffs render with truecolor
backgrounds
- [x] On Windows Terminal with `FORCE_COLOR` set: confirm promotion is
disabled and output follows the forced `supports-color` level
- [x] On macOS/Linux terminals: confirm no behavior change
Fixes https://github.com/openai/codex/issues/12904
Fixes https://github.com/openai/codex/issues/12890
Fixes https://github.com/openai/codex/issues/12912
Fixes https://github.com/openai/codex/issues/12840
## Why
`PermissionProfile` should describe filesystem roots as absolute paths
at the type level. Using `PathBuf` in `FileSystemPermissions` made the
shared type too permissive and blurred together three different
deserialization cases:
- skill metadata in `agents/openai.yaml`, where relative paths should
resolve against the skill directory
- app-server API payloads, where callers should have to send absolute
paths
- local tool-call payloads for commands like `shell_command` and
`exec_command`, where `additional_permissions.file_system` may
legitimately be relative to the command `workdir`
This change tightens the shared model without regressing the existing
local command flow.
## What Changed
- changed `protocol::models::FileSystemPermissions` and the app-server
`AdditionalFileSystemPermissions` mirror to use `AbsolutePathBuf`
- wrapped skill metadata deserialization in `AbsolutePathBufGuard`, so
relative permission roots in `agents/openai.yaml` resolve against the
containing skill directory
- kept app-server/API deserialization strict, so relative
`additionalPermissions.fileSystem.*` paths are rejected at the boundary
- restored cwd/workdir-relative deserialization for local tool-call
payloads by parsing `shell`, `shell_command`, and `exec_command`
arguments under an `AbsolutePathBufGuard` rooted at the resolved command
working directory
- simplified runtime additional-permission normalization so it only
canonicalizes and deduplicates absolute roots instead of trying to
recover relative ones later
- updated the app-server schema fixtures, `app-server/README.md`, and
the affected transport/TUI tests to match the final behavior
## Why
The `notify` hook payload did not identify which Codex client started
the turn. That meant downstream notification hooks could not distinguish
between completions coming from the TUI and completions coming from
app-server clients such as VS Code or Xcode. Now that the Codex App
provides its own desktop notifications, it would be nice to be able to
filter those out.
This change adds that context without changing the existing payload
shape for callers that do not know the client name, and keeps the new
end-to-end test cross-platform.
## What changed
- added an optional top-level `client` field to the legacy `notify` JSON
payload
- threaded that value through `core` and `hooks`; the internal session
and turn state now carries it as `app_server_client_name`
- set the field to `codex-tui` for TUI turns
- captured `initialize.clientInfo.name` in the app server and applied it
to subsequent turns before dispatching hooks
- replaced the notify integration test hook with a `python3` script so
the test does not rely on Unix shell permissions or `bash`
- documented the new field in `docs/config.md`
## Testing
- `cargo test -p codex-hooks`
- `cargo test -p codex-tui`
- `cargo test -p codex-app-server
suite::v2::initialize::turn_start_notify_payload_includes_initialize_client_name
-- --exact --nocapture`
- `cargo test -p codex-core` (`src/lib.rs` passed; `core/tests/all.rs`
still has unrelated existing failures in this environment)
## Docs
The public config reference on `developers.openai.com/codex` should
mention that the legacy `notify` payload may include a top-level
`client` field. The TUI reports `codex-tui`, and the app server reports
`initialize.clientInfo.name` when it is available.
- replace show_nux with structured availability_nux model metadata
- expose availability NUX data through the app-server model API
- update shared fixtures and tests for the new field
Addresses bug https://github.com/openai/codex/issues/12589
Builds on community PR #12763.
This adds `oauth_resource` support for MCP `streamable_http` servers and
wires it through the relevant config and login paths. It fixes the bug
where the configured OAuth resource was not reliably included in the
authorization request, causing MCP login to omit the expected
`resource` parameter.
## Summary
This changes `custom_tool_call_output` to use the same output payload
shape as `function_call_output`, so freeform tools can return either
plain text or structured content items.
The main goal is to let `js_repl` return image content from nested
`view_image` calls in its own `custom_tool_call_output`, instead of
relying on a separate injected message.
## What changed
- Changed `custom_tool_call_output.output` from `string` to
`FunctionCallOutputPayload`
- Updated freeform tool plumbing to preserve structured output bodies
- Updated `js_repl` to aggregate nested tool content items and attach
them to the outer `js_repl` result
- Removed the old `js_repl` special case that injected `view_image`
results as a separate pending user image message
- Updated normalization/history/truncation paths to handle multimodal
`custom_tool_call_output`
- Regenerated app-server protocol schema artifacts
## Behavior
Direct `view_image` calls still return a `function_call_output` with
image content.
When `view_image` is called inside `js_repl`, the outer `js_repl`
`custom_tool_call_output` now carries:
- an `input_text` item if the JS produced text output
- one or more `input_image` items from nested tool results
So the nested image result now stays inside the `js_repl` tool output
instead of being injected as a separate message.
## Compatibility
This is intended to be backward-compatible for resumed conversations.
Older histories that stored `custom_tool_call_output.output` as a plain
string still deserialize correctly, and older histories that used the
previous injected-image-message flow also continue to resume.
Added regression coverage for resuming a pre-change rollout containing:
- string-valued `custom_tool_call_output`
- legacy injected image message history
#### [git stack](https://github.com/magus/git-stack-cli)
- 👉 `1` https://github.com/openai/codex/pull/12948
jif-oai ·
pash-openai ·
Celia Chen ·
Brian Fioca ·
Ahmed Ibrahim ·
Ruslan Nigmatullin ·
Dylan Hurd ·
Eric Traut ·
daveaitel-openai ·
Leo Shimonaka ·
xl-openai ·
Michael Bolin ·
Andi Liu ·
alexsong-oai ·
Charley Cunningham ·
Felipe Coury ·
Curtis 'Fjord' Hawthorne ·