## What changed
- Add `x-cluster-name` from `OPENAI_CLUSTER` and
`x-openai-internal-caller` from `DD_SERVICE` to rendezvous WebSocket
handshakes when the environment values are nonempty and valid HTTP header
values.
- Apply the rendezvous headers to both client and remote executor connections
while preserving the existing W3C trace context headers.
GitOrigin-RevId: fead0244ba24fa208990b92d2170a9ae76ff2ceb
## Why
Image generation did not save its result when the extension host did not provide a host-local save root.
## What changed
- Save generated images under the turn environment's `generated_images` directory through its sandboxed filesystem executor when no host save root is configured.
- Limit executor-written images to 32 MiB, reject symlinked output directories and existing destinations, and continue returning the image if persistence fails.
- Include the saved workspace path in the tool output hint while preserving host-managed persistence when a save root is available.
## Testing
- Extend the granted-turn-permissions integration test to verify the image and path hint are written to the turn workspace without creating a host-local artifact.
GitOrigin-RevId: 3d24381a75fe4595641b5250e19fac8ed473b2cd
## What changed
- Stop parsing and exposing the `model` field from skill frontmatter.
- Remove the skill model delegation types and instruction generation.
GitOrigin-RevId: 01198c68c095da5062e8abec417cc3539099d9d5
## What changed
- Add a `desktop.security.enforcement` doctor check for macOS that assesses the app with Gatekeeper and classifies recent Gatekeeper and XProtect events.
- Add the same check on Windows by inspecting recent Microsoft Defender, AppLocker, and Windows App Control events for Codex executables.
- Report blocked or quarantined executions as failures, audit-only or unavailable evidence as warnings, and include actionable remediation while bounding and redacting collected event details.
## Testing
- Add coverage for platform event classification, trusted executable matching, unavailable diagnostics, remediation, and bounded redacted evidence.
GitOrigin-RevId: 792844390cd2cf92d3bc20e6a0973020b4364e51
## What changed
- Skip hyperlink layout for wrapped rows above the scroll offset and stop once
processing reaches the bottom of the viewport.
- Resolve each rendered link destination once while marking its visible cells.
- Add coverage for hyperlinks that span visible wrapped rows after scrolling.
GitOrigin-RevId: c1506d47fa3f1b895966dc2cde7d5b859c201792
## What changed
- Stop treating Vim normal-mode history-up as an alias for editing the latest queued message.
- Continue to edit queued messages through the configured `chat.edit_queued_message` binding.
GitOrigin-RevId: d830e0976a38725e948d420c1a5f374bb13fe945
## Why
Scrolling the transcript pager previously rendered a scratch buffer containing
all rows through the visible viewport, including hidden content above it. This
made rendering work and buffer size grow with the scroll offset.
## What changed
- Add viewport-aware rendering for committed transcript cells and the live tail.
- Forward scrolled rendering through cached and inset renderables while keeping
the full-height fallback for renderables that do not support it.
- Keep hyperlink metadata aligned with scrolled text.
## Testing
Add coverage comparing viewport-aware output with the full-height fallback
across wrapping widths, offsets, insets, styles, hyperlinks, and live tails.
GitOrigin-RevId: 630c30c4a61969a50f2b34701ca1dd87788a4298
## Why
The streaming TUI rerendered the entire mutable Markdown block whenever a
newline-bearing delta arrived. For long fenced code blocks, this repeatedly
parsed and syntax-highlighted the growing fence.
## What changed
- Detect open, top-level, language-tagged code fences and append newly
completed lines directly to the existing render.
- Preserve Syntect parser state across appended lines so multiline syntax and
the active theme remain consistent.
- Fall back to the canonical Markdown renderer for ambiguous fence syntax,
theme changes, incomplete lines, and highlighting limit transitions.
## Testing
Added coverage that compares streamed output with full renders across syntax
styles, Unicode, links, blank lines, fence variants, theme changes, and
highlighting limits. A long-fence test also verifies that the rendered prefix
is retained as new lines arrive.
GitOrigin-RevId: 036db2b7bc93607083b8001f8b4fa3cbb58c713e
## What changed
- Detect installed Codex desktop apps on macOS and Windows and report their version, running state, and redacted log location.
- Inspect bounded log data for the active desktop session and report the latest local app-server handshake outcome without exposing other log fields.
- Add a dedicated Desktop App section to human-readable doctor reports.
## Testing
- Cover local versus remote handshakes, failed handshakes with sensitive fields, bounded log reads, and stopped desktop sessions.
GitOrigin-RevId: ecfb4bd7af3589263624771ec9413b51953280eb
## What changed
Add `server_kind=openai_codex_apps` to the
`codex.mcp.protocol_discovery` counter and its duration metric when protocol
discovery runs for the Codex Apps MCP server. Other MCP servers retain the
existing `mode` and `outcome` tags.
GitOrigin-RevId: aadbdc54642f2c103a9f1bab39308e962c115f75
## Why
Recording a late frame's requested deadline as its emission time can leave the
rate limiter with a stale timestamp, allowing the next frame to be emitted
immediately in a redraw burst.
## What changed
Record the current time when the frame scheduler sends a draw notification so
the 120 FPS limit is measured from the actual emission.
## Testing
Add a paused-time regression test that submits an overdue draw and verifies the
next frame waits for `MIN_FRAME_INTERVAL`.
GitOrigin-RevId: 8c9a5ac0bf858b2ad87838f6d7c67c9470fdfc61
## What changed
- Add `EnvironmentNetworkPolicy` as a serializable view of portable domain,
Unix-socket, upstream-proxy, and local-binding restrictions without exposing
controller-owned proxy runtime settings.
- Add an optional `network_policy` field to `EnvironmentConfig` and re-export
the policy and related permission types through the core APIs.
- Reject owner-provided network policies during environment preview and
readiness until runtime enforcement is implemented, while preserving the
existing controller policy when the field is absent.
## Testing
- Cover rejection through both environment preview and readiness, including
preservation of the existing environment selection.
GitOrigin-RevId: 80ad4cf4d4a45632daa7a823e6cf568eb0e8bb80
## Why
The app directory is a discovery catalog and can include apps that are not
callable in the active thread. Using it for `$` suggestions can therefore
offer mentions that cannot be used.
## What changed
- Load mention candidates from `app/installed`, enrich callable apps through
bounded `app/read` batches, and retain compatibility with app servers that
do not support `app/installed`.
- Keep the mention catalog separate from the `/apps` directory cache and scope
results to the current thread, workspace, and account.
- Refresh mentions after app notifications, enablement changes, configuration
updates, and apps MCP readiness, while immediately removing disabled apps.
## Testing
Add coverage for batching and fallback behavior, stale scoped results,
notification refreshes, enablement changes, MCP startup races, and filtering
non-callable apps from mentions.
GitOrigin-RevId: 4d9e20794aed30cc97120170792fb555b86fe1e7
## Why
Creating the telemetry shutdown thread during shutdown can fail under resource
pressure, including when the native thread guard page cannot be allocated.
## What changed
- Start and verify a dedicated shutdown worker when `OtelProvider` is created.
- Send the provider to the prepared worker for bounded shutdown, while preserving
timeout behavior and avoiding a potentially blocking destructor if worker
preparation failed.
- Rename the fallible provider constructor to `try_new`.
## Testing
Add Unix regression coverage that injects guard-page allocation failures, plus
coverage for worker preparation failure, successful shutdown, and timeouts.
GitOrigin-RevId: 3656298078a800a7fa392437c2ee4a68753092e3
## Why
The app server sends sub-agent activity in both item-started and item-completed
notifications, causing the TUI to add the same activity to history twice.
## What changed
Render live sub-agent activity when the item completes instead of also rendering
it when the item starts. Preserve the existing replay behavior.
## Testing
Add a regression test that sends matching started and completed notifications
and verifies that the activity produces one history cell.
GitOrigin-RevId: 76a66f3657fe297f37067ed672ab9642997b4499
## What changed
Gate plugin hook resolution in `hooks/list` on both the hooks and plugins
feature flags. When hooks are disabled, return the default empty plugin hook
outcome without loading plugin hook sources or warnings.
GitOrigin-RevId: 166070576effaba248b790c4f2a47e6e9a7b1112
## Why
MCP requests can contain sensitive headers and tool-call bodies. Following a
cross-origin redirect could disclose them to another server.
## What changed
- Follow redirects only when every hop remains on the configured server's
origin, and require HTTPS for redirects on non-loopback hostnames.
- Preserve standard redirect method and body handling while sharing the
original timeout and enforcing a 10-hop limit.
- Avoid replaying plaintext proxy credentials across redirects, while leaving
OAuth requests that stop at redirects unchanged.
## Testing
Add coverage for same-origin and cross-origin redirects, redirect methods,
headers and bodies, proxy credentials, timeouts, hop limits, and legacy MCP
tool calls.
GitOrigin-RevId: 3e267911a1665ad8c2fc63825099a758770f4d68
## What changed
- Register `in_app_chat` and `in_app_dictation` as stable, default-enabled feature gates for desktop apps.
- Allow requirements to disable either capability and expose both keys in the generated configuration schema.
## Testing
- Add configuration tests that verify enterprise requirements can disable each gate.
GitOrigin-RevId: 5bba35d710fb0b6924aa55df98b2e013fa288f6d
## What changed
- Add `cli_auth_credentials_store` and `chatgpt_base_url` as exact managed requirements loaded from local requirements configuration.
- Apply these requirements to runtime and bootstrap authentication configuration, overriding user-configured values.
- Expose the managed values through `configRequirements/read` and reject attempts to change them through config write APIs.
- Ignore these local-only authentication requirements in cloud-managed requirement layers.
## Testing
- Cover requirement loading and composition, runtime and bootstrap overrides, app-server responses, and read-only config writes.
GitOrigin-RevId: 8ea9c096386072036ca5441f3d48af8707988a65
## Why
Unified exec could drop network approval registration when a selected environment used a Windows working directory on a non-Windows host because the trigger path could not be converted to a host absolute path.
## What changed
- Keep network approval trigger working directories as `PathUri` values and render them with their inferred native path syntax for Guardian requests.
- Fall back to the selected turn environment only when a trigger path can be converted to a host absolute path.
- Reject selected environment working directories larger than 8 KiB at the thread-manager validation boundary.
## Testing
- Cover managed-network approval registration for a foreign Windows working directory.
- Verify that oversized working directories are rejected as invalid `thread/start` requests.
GitOrigin-RevId: 646fbf12115a35a7e5cb04b511464797bee6647b
## Why
Resumed and forked threads need their persisted Guardian risk state available when routing later approval requests.
## What changed
- Initialize a session's `SecurityRiskScore` from the latest score in its rollout history when no score was supplied explicitly.
- Preserve the inherited score when resuming or forking a thread so low-risk actions can skip redundant Guardian reviews.
## Testing
- Add app-server coverage for inheriting the latest Guardian score when resuming and forking threads.
GitOrigin-RevId: 95a6c803f8dae75516d1da9d66a722379a0a7ef2
## Why
Config reads arriving while a long-running shared read was active waited for that
request to finish, even though they could safely run concurrently.
## What changed
- Admit newly queued shared reads into the active batch until an exclusive write
is queued, preserving the write as a fairness barrier.
- Treat `hooks/list` and `configRequirements/read` as shared config reads.
- Cover concurrent reads, read/write ordering, and config reads completing
alongside a `skills/list` request.
GitOrigin-RevId: 8194301c2ba60be0fc71b8915923af9accc745dc
## What changed
- Add end-to-end app-server tests for MCP tool approvals with Guardian V2 and automatic review enabled.
- Verify that sampling receives the thread context, uses a thread-scoped prompt cache key, and persists the resulting action risk score.
- Cover both routing outcomes: low-risk actions skip later Guardian reviews, while high-risk actions continue through full review.
GitOrigin-RevId: 78ba2b0560caf701ad333785c4dda4f7d2be4f2c
## Why
Durable thread queues can be updated through another SQLite connection, but loaded idle threads need to notice those updates before they can dispatch the new messages.
## What changed
- Track a durable revision for each thread queue, including backfilling existing queues and updating revisions on inserts, updates, and deletes.
- Poll SQLite's data version and query revisions to find changed queues for loaded threads.
- Wake idle threads with pending external messages, discover queued work when threads are loaded or resumed, and retry each thread independently so one blocked queue does not stall others.
## Testing
- Cover cross-runtime queue writes, edits, independent dispatch, wake retries, and resumed threads.
- Cover migration backfills and revision tracking for queue updates and deletes.
GitOrigin-RevId: 906d902bbf6a760be206a1b2c08fd71427b63c46
## Why
Building a resume-picker preview for a legacy session should not require loading its full thread history.
## What changed
- Read up to six preview lines by scanning at most 1 MiB from the end of a local legacy rollout.
- Fall back to the existing history hydration path when the local file is unavailable, the bounded scan is incomplete, or the tail contains a rollback.
- Keep the initial `thread/read` request turn-free while loading the preview.
## Testing
Added coverage for compaction, rollbacks, oversized records, exhausted scan budgets, and app-server request parameters.
GitOrigin-RevId: a8950f03d8b56323f268fd3355c6a38a16cb9833
## Why
Direct access to shimmer primitives can bypass the reduced-motion handling
introduced in https://github.com/openai/codex/pull/20564.
## What changed
- Make `shimmer` a private child of `motion` and restrict `shimmer_spans` to
its parent module, so Rust visibility enforces the animation boundary.
- Remove the source-scanning regression test that previously checked this
boundary with regular expressions.
GitOrigin-RevId: 1f4cccacaa907992c636cd1e9cd2829b7653f811
## What changed
- Reuse shared directory fixtures and local and remote app-server instances across the resume cwd case matrix.
- Seed the runtime working-directory override directly so each case focuses on resume behavior without starting and replacing an extra thread.
- Run the test on a four-thread Tokio runtime and shut down each shared server after all cases complete.
GitOrigin-RevId: 4bc4978abc546882f26bcb5e3591ff9673758b4a
## Why
Pet tests reuse the same blank WebP spritesheet fixture, so encoding a new image
for every temporary path does unnecessary repeated work.
## What changed
Encode the test spritesheet once with `LazyLock`, then write the cached WebP bytes
to each requested fixture path.
GitOrigin-RevId: 26d75bb1bac448717ca63aeb82eb4e500c86fad8
## Why
The 1,000-character human-like typing test waited on the wall clock after
every keypress, making the test unnecessarily slow.
## What changed
Advance an injected timestamp by the recommended paste flush delay and flush
the paste-burst detector explicitly after each character. This preserves the
coverage that slow typing appears directly without creating a paste
placeholder, without sleeping between keypresses.
GitOrigin-RevId: 0c1beb40283f7d73a381cd221725ab4227c8e233
## What changed
Construct the oversized unified diff fixture directly instead of generating it
from two large input strings. Keep the fixture just above
`MAX_HIGHLIGHT_LINES` so the test continues to verify that syntax highlighting
is skipped for large updates.
GitOrigin-RevId: 8f8853e03a50ff9f53e77b2ba0956cf85ee17c07
## Why
MCP tool results can contain multi-megabyte image, audio, or resource bodies,
even though TUI history renders only summaries for those content types.
## What changed
- Project completed MCP results into width-independent display data, dropping
media and resource bodies that history does not render.
- Preserve text output, JSON fallbacks, error state, Node REPL rendering, and
the fully decoded image-output marker.
- Add tests covering width-dependent rendering, image validation, and Node REPL
text fields on non-text and unknown content blocks.
GitOrigin-RevId: 2cfddf0864093e6837a4431e541f7a9cab544a68
## Why
Keeping overflowing separators with the following word can add whitespace-only
rows and make composer wrapping differ from the queued-input preview.
## What changed
- Hang breakable whitespace from the preceding visual row at soft word breaks
while preserving the underlying editable text.
- Keep indentation, trailing whitespace, mandatory breaks, and nonbreaking spaces
visible and editable according to their existing semantics.
- Align cursor placement, scrolling, vertical navigation, masked rendering, and
styled overlays with the visible wrapped content, including after resizes.
## Testing
Add focused wrapping, rendering, cursor, editing, scrolling, navigation, Unicode
separator, long whitespace-run, and resize coverage.
GitOrigin-RevId: 7fbdb0a4bf33cfee20c4c19dca5aee74df4f892c
## Why
Windows named pipe servers can impersonate connected clients. IDE context
connections only require the server to identify the client, so they should not
grant broader impersonation rights.
## What changed
Open Windows IDE context pipes with `SECURITY_SQOS_PRESENT` and
`SECURITY_IDENTIFICATION`, limiting the server's impersonation token to the
identification level.
## Testing
Add a Windows regression test that connects to a named pipe, asks the server to
impersonate the client, and verifies that the resulting token has the
`SecurityIdentification` level.
GitOrigin-RevId: 51ef080f3367857987e03b0113a476040ea40e56
## Why
Headers configured for an MCP resource can contain credentials and must not be
forwarded to a different origin during OAuth discovery, registration, or token
exchange.
## What changed
- Apply configured resource headers only to OAuth requests on the MCP resource
origin, while allowing OAuth-generated headers to take precedence.
- Follow credentialed OAuth redirects only when each destination remains on the
resource origin. Preserve redirect method semantics, the overall timeout, the
response size limit, and a maximum of 10 redirects.
- Use the Codex MCP user agent for requests to authorization-server origins.
## Testing
Add coverage for cross-origin header isolation, same-origin dynamic client
registration redirects, OAuth header precedence, token refresh, timeout
accounting, and response size limits.
GitOrigin-RevId: 609d06e2b065e4aa9ca5382b4ae60455f9e26db9
## What changed
- Define MCP server identities, matchers, requirements, and per-plugin MCP requirements in the new `codex_protocol::mcp_policy` module.
- Keep requirement validation and matching against `McpServerConfig` in `codex-config`.
- Re-export the moved types from `codex-config` to preserve existing consumers.
GitOrigin-RevId: ad8a107e3085191241a5766f061dd004103e21e6
## Why
Short continuation prompts such as `continue` do not contain enough context for
skill selection on their own.
## What changed
- Add the `task_context_fusion_v1` shadow selector, combining the current request
with up to two prior substantive requests and recently relevant skills.
- Record explicit skill intent and successful skill invocations for future turns,
while excluding same-turn observations from predictions.
- Bound retained requests, augmented queries, and skill history, including safe
truncation at UTF-8 character boundaries.
## Testing
Add unit and extension tests for continuation prompts, explicit intent, turn
isolation, cold thread state, bounded history, and unchanged control selectors.
GitOrigin-RevId: 72eca3f0d64620a0d24e95d5675f126ac982f8c7
## Why
Guardian stores a `SecurityRiskScore` before advancing its scored-tool-call
counter. Waiting only for the score can therefore let the test continue while
the asynchronous sampler is still updating its progress state.
## What changed
Wait for `latest_scored_tool_call` to advance before exercising Guardian v2
approval decisions, and reuse the same progress tracker for the subsequent
assertions.
GitOrigin-RevId: 3d92c9a3ccf52f4f43a2454aac60d2479258c1c8
## What changed
- Track app-server turn lifecycle events for OpenAI API-key sessions when OTLP telemetry export is enabled.
- Query the turn-cost endpoint after completed or aborted turns and emit `codex.turn_cost` with the estimated USD cost, interruption state, model, speed, and reasoning effort.
- Preserve provider organization and project headers, wait for all observed responses to be priced, and retry pending or transiently failed queries with bounded tracking.
- Re-probe availability after authentication changes and shut the worker down with other app-server background tasks.
## Testing
- Cover late API-key login, transient probe failures, telemetry after thread removal, and turns with multiple responses.
- Verify turn-cost requests use API-key authentication and provider scope headers and deserialize priced response data.
GitOrigin-RevId: a303162a54798b554fcb2e9e2b372e1cf9f49f39
## Why
Guardian V2 should not override a managed approval-reviewer policy that excludes
the user reviewer.
## What changed
- Force `features.guardianv2` off when `allowed_approvals_reviewers` does not
include `user`, for both local requirements and cloud config bundles.
- Preserve Guardian V2 when the user reviewer remains allowed or when only
legacy Guardian feature settings are present.
- Return the effective disabled value through `config/read`.
## Testing
Added coverage for cloud requirements and the app-server config RPC.
GitOrigin-RevId: 82b6d0aaf47fbaa58ae349d4e56777cd442f5807
## What changed
- Track the latest tool call and latest scored tool call for each Guardian v2 thread.
- Skip approval review when the score lags by more than `max_tool_call_lag` tool calls.
- Add `max_tool_call_lag` to the Guardian v2 configuration and schema, with a default of three.
## Testing
- Cover approval review at, above, and after recovering from the configured lag limit.
- Cover deserialization of `max_tool_call_lag`.
GitOrigin-RevId: fd870e143d95ee76491180e2aed742717248ff78
## Why
Guardian sampling can encounter temporary connection or service failures while
opening a WebSocket, starting a request, or consuming its response stream.
## What changed
- Retry transient transport, streaming, rate-limit, and server failures up to
two times across the full sampling flow.
- Recover from unauthorized responses through the configured auth manager and
discard pooled connections that may contain stale credentials.
- Allow sampler initialization to continue when its initial WebSocket prewarm
fails, so the first sample can establish a fresh connection.
## Testing
Added coverage for recovery after prewarm failures and `503` responses, and for
stopping once the transient retry budget is exhausted.
GitOrigin-RevId: af74c2df95c4241184ca641ef4418db6aa4ab695
## What changed
- Stop fetching the ChatGPT workspace `enable_plugins` setting.
- Let the effective Codex feature configuration control app and plugin APIs,
plugin-provided skills and hooks, and experimental feature reporting.
GitOrigin-RevId: 46aebf0f07cd47e093e48663e41207cdf9a38c9f
## What changed
- Add shadow selectors that fuse recently invoked skills with character routing-card matches using reciprocal rank fusion.
- Add a second variant that also incorporates weighted lexical matches.
- Record both variants in the existing shadow-selection experiment, capped at 50 candidates.
## Testing
- Cover cold starts, duplicate removal, short exact matches, deterministic ranking, result limits, and truncation metadata.
- Extend the skills integration test to verify both selectors recover a skill invoked on an earlier turn.
GitOrigin-RevId: 4ff9e91da8cb49ecbce54ab8e760820616e1a1d3
## What changed
- Add optional `guardian_v2` model-message defaults for classifier instructions,
review thresholds, reasoning effort, transcript selection and limits, and
action and compaction token limits.
- Apply those defaults when Guardian v2 samples a tool call, while preserving
explicit `[features.guardianv2]` settings over catalog values.
- Preserve the new catalog configuration when applying model overrides.
## Testing
- Cover model-catalog serialization and model override behavior.
- Verify that sampling uses catalog defaults while retaining local overrides.
GitOrigin-RevId: 4ae9ec235c647acbca05a20b7c4b87a4e1331161
## What changed
- Add the `features.guardianv2.transcript.include_images` opt-in setting.
- Include recent images from user and assistant messages and configured tool outputs in Guardian v2 sampling requests.
- Bound image context to the four most recent images and 8 MiB total, and omit image detail hints when constructing the sampling request.
## Testing
- Cover configuration parsing and image forwarding from messages and tool outputs.
GitOrigin-RevId: b0ae6b6182535fd084ea103ed01dc3c803d7d1f7
## What changed
- Add `max_parent_compaction_tokens`, defaulting to 25,000 tokens, to the Guardian v2 feature configuration.
- Reuse the latest encrypted parent compaction only when its complete serialized item fits within the configured limit.
- Fail closed with the maximum action-risk score when the latest encrypted compaction is oversized, instead of sampling with older context.
## Testing
- Cover exact-boundary and oversized compaction items, including oversized passthrough metadata.
- Verify that an oversized latest compaction bypasses sampling and prevents approval.
GitOrigin-RevId: 0b308e7bec0f41d46433dfdf656948cee5cba667
## What changed
- Add `[skills].max_context_tokens` to override the token budget used to render the available-skills catalog.
- Require a positive value and cap configured budgets at 10,000 tokens.
- Preserve the existing default of 2% of the model context window when the option is unset.
## Testing
- Cover configuration parsing, override behavior, the 10,000-token cap, and catalog rendering within the configured budget.
GitOrigin-RevId: f29ddcb57fc2d800c0006d03e1c33cbe3ab6272d
## What changed
- Ignore the default `CODEX_HOME/managed_config.toml` on Windows and exclude it
from local managed-configuration detection.
- Emit a startup warning when the deprecated file exists, directing users to
`%ProgramData%\OpenAI\Codex\requirements.toml` for enforced settings or
`config.toml` for defaults.
- Preserve explicit managed-config path overrides and Unix legacy-file support.
## Testing
- Add Windows tests covering ignored legacy settings, the startup warning, and
managed-configuration detection through `requirements.toml`.
GitOrigin-RevId: a61d9d9912b13817ba82807a486c9ed92e49c7bc
## What changed
- Add an optional restrictive execution policy to `EnvironmentConfig` and merge it over the active command policy for shell execution and approval decisions.
- Reject environment policies that contain allow rules, so environment configuration can only tighten command access.
- Include the environment policy fingerprint in cached approval keys so policy changes require fresh session approvals.
## Testing
- Cover restrictions overriding saved prefix approvals, policy changes invalidating session approvals, and environment overlays remaining effective when model-specific prefix filtering is active.
GitOrigin-RevId: e94f877549ed5e0f577b8cb52f3c1872c58f029d
## Why
`CODEX_EXEC_SERVER_NOISE_AUTH_TOKEN` is an execution-server credential and
must not be exposed to model-reachable commands or command hooks, including
when shell environment policy explicitly sets a case variant of the name.
## What changed
- Add `CODEX_EXEC_SERVER_NOISE_AUTH_TOKEN` to the shared list of
non-inheritable environment variables.
- Keep the environment variable constant in `codex-protocol` so the exec
server and environment scrubber use the same definition.
- Verify case-insensitive removal after policy overrides for shell commands
and command hooks.
GitOrigin-RevId: 759b224b6a7fb4f56f7b1a6d94ffbb0b188c658f
