Won Park 0dbad2a348 Make auto-review denial short-circuit use a rolling review window (#22110) ...

## Why

Long-running turns can accumulate enough denied auto-review decisions to
trip the global short-circuit even when those denials are spread far
apart. The breaker should still stop genuinely bad loops, but it should
judge recent behavior instead of lifetime turn history.

## What changed

- Replaced the lifetime `10 total denials` threshold with `10 denials in
the last 50 reviews`.
- Kept the existing `3 consecutive denials` interrupt behavior
unchanged.
- Tracked recent auto-review outcomes in the circuit breaker and updated
the warning copy to report the rolling-window count.
- Renamed the new rolling-window coverage to `auto_review_*` test names.
- Added coverage that confirms older denials fall out of the 50-review
window and no longer trigger the breaker.

## Validation

- `just fmt`
- `cargo test -p codex-core guardian_rejection_circuit_breaker --lib`
- `cargo test -p codex-core auto_review_rejection_circuit_breaker --lib`

2026-05-11 11:03:11 -07:00

..

agent

[codex] Remove remote thread store implementation (#21596)

2026-05-08 00:02:46 +00:00

apps

Split DeveloperInstructions into individual fragments. (#18813)

2026-04-21 10:22:36 -07:00

bin

Refactor config types into a separate crate (#16962)

2026-04-07 00:32:41 +00:00

config

chore: drop built-in MCPs (#22173)

2026-05-11 19:45:08 +02:00

context

Improve goal continuation based on feedback (#22045)

2026-05-11 09:51:21 -07:00

context_manager

Avoid hard-coded environment context shell (#21390)

2026-05-06 19:54:26 +00:00

guardian

Make auto-review denial short-circuit use a rolling review window (#22110)

2026-05-11 11:03:11 -07:00

plugins

[codex] Remove string-keyed MCP tool maps (#21454)

2026-05-07 10:16:10 -07:00

sandboxing

permissions: derive legacy exec policies at boundaries (#19737)

2026-04-26 22:11:49 -07:00

session

Fix goal update and add /goal edit command in TUI (#21954)

2026-05-11 10:49:19 -07:00

state

extension: wire extension registries into sessions (#21737)

2026-05-11 11:38:18 +02:00

tasks

[codex-analytics] add item lifecycle timing (#20514)

2026-05-04 22:33:20 +00:00

tools

feat: move extensions tool (#22163)

2026-05-11 17:14:43 +02:00

unified_exec

Remove ToolName display helper (#21465)

2026-05-08 12:17:48 -07:00

utils

Extract rollout into its own crate (#15548)

2026-03-24 18:10:53 -07:00

agents_md_tests.rs

Remove js_repl feature (#19410)

2026-04-24 17:49:29 -07:00

agents_md.rs

[codex] Move config loading into codex-config (#19487)

2026-04-26 15:10:53 -07:00

apply_patch_tests.rs

[codex] Migrate apply_patch to executor filesystem (#17027)

2026-04-07 21:20:22 +00:00

apply_patch.rs

Make turn diff tracking operation backed (#21180)

2026-05-07 11:33:47 +02:00

arc_monitor_tests.rs

Surface multi-environment choices in environment context (#20646)

2026-05-01 22:11:06 +00:00

arc_monitor.rs

feat: add remote compaction v2 Responses client path (#20773)

2026-05-04 14:15:01 +02:00

attestation.rs

[codex] request desktop attestation from app (#20619)

2026-05-08 12:36:02 -07:00

client_common_tests.rs

Allow guardian bare allow output (#18797)

2026-04-21 15:37:12 -07:00

client_common.rs

[codex] Trace cancelled inference streams (#19839)

2026-04-27 21:58:29 +00:00

client_tests.rs

[codex] request desktop attestation from app (#20619)

2026-05-08 12:36:02 -07:00

client.rs

Add x-codex-ws-stream-request-start-ms (#22113)

2026-05-11 08:15:52 -07:00

codex_delegate_tests.rs

[codex-analytics] plumb protocol-native review timing (#21434)

2026-05-07 20:31:41 -07:00

codex_delegate.rs

extension: wire extension registries into sessions (#21737)

2026-05-11 11:38:18 +02:00

codex_thread.rs

Reapply "Move skills watcher to app-server" (#21652)

2026-05-08 17:41:15 -07:00

command_canonicalization_tests.rs

fix: move inline codex-rs/core unit tests into sibling files (#14444)

2026-03-12 08:16:36 -07:00

command_canonicalization.rs

core: remove cross-crate re-exports from lib.rs (#16512)

2026-04-01 23:06:24 -07:00

compact_remote_v2.rs

Send response.processed after remote compaction v2 (#21642)

2026-05-07 19:57:36 -07:00

compact_remote.rs

Omit service_tier from remote /responses/compact requests under API auth (#21676)

2026-05-08 11:15:14 +03:00

compact_tests.rs

Revert "[codex] Harden overflow auto-compaction recovery" (#22170)

2026-05-11 19:33:15 +03:00

compact.rs

Revert "[codex] Harden overflow auto-compaction recovery" (#22170)

2026-05-11 19:33:15 +03:00

config_lock.rs

feat: export and replay effective config locks (#20405)

2026-05-01 17:46:02 +02:00

connectors_tests.rs

Using cached connector directory for discoverable tools list (#21497)

2026-05-08 14:14:11 -07:00

connectors.rs

Using cached connector directory for discoverable tools list (#21497)

2026-05-08 14:14:11 -07:00

consequential_tool_message_templates.json

[elicitation] User-friendly tool call messages. (#14403)

2026-03-12 00:35:21 -07:00

environment_selection.rs

Make environment provider snapshots path-free (#21794)

2026-05-08 15:30:00 -07:00

event_mapping_tests.rs

Improve goal continuation based on feedback (#22045)

2026-05-11 09:51:21 -07:00

event_mapping.rs

Organize context fragments (#18794)

2026-04-20 22:39:17 -07:00

exec_env_tests.rs

[codex] Move config loading into codex-config (#19487)

2026-04-26 15:10:53 -07:00

exec_env.rs

[codex] Move config loading into codex-config (#19487)

2026-04-26 15:10:53 -07:00

exec_policy_tests.rs

fix(exec_policy) heredoc parsing file_redirect (#20113)

2026-05-01 01:05:02 +00:00

exec_policy_windows_tests.rs

execpolicy: unwrap PowerShell -Command wrappers on Windows (#20336)

2026-05-01 00:56:20 +00:00

exec_policy.rs

fix(exec_policy) heredoc parsing file_redirect (#20113)

2026-05-01 01:05:02 +00:00

exec_tests.rs

fix: handle deferred network proxy denials (#19184)

2026-04-29 19:13:57 +00:00

exec.rs

fix: handle deferred network proxy denials (#19184)

2026-04-29 19:13:57 +00:00

flags.rs

…

function_tool.rs

nit: drop unused function call error (#8903)

2026-01-08 15:07:30 +00:00

git_info_tests.rs

Refactor config loading to use filesystem abstraction (#18209)

2026-04-17 00:51:21 +00:00

goals.rs

Fix goal update and add /goal edit command in TUI (#21954)

2026-05-11 10:49:19 -07:00

hook_runtime.rs

Add compact lifecycle hooks (started by vincentkoc - external contrib) (#19905)

2026-05-06 18:08:31 -07:00

installation_id.rs

feat: split memories part 2 (#19860)

2026-04-28 13:03:28 +02:00

landlock.rs

linux-sandbox: switch helper plumbing to PermissionProfile (#20106)

2026-04-28 19:43:44 -07:00

lib.rs

extension: move git attribution into an extension (#21738)

2026-05-11 12:53:15 +02:00

mcp_openai_file.rs

refactor: route Codex auth through AuthProvider (#18811)

2026-04-23 17:14:02 -07:00

mcp_skill_dependencies.rs

feat: make built-in MCPs first-class runtime servers (#21356)

2026-05-07 10:36:32 +02:00

mcp_tool_approval_templates.rs

chore: clean up argument-comment lint and roll out all-target CI on macOS (#16054)

2026-03-27 19:00:44 -07:00

mcp_tool_call_tests.rs

Route opted-in MCP elicitations through Guardian (#19431)

2026-05-06 19:42:45 +00:00

mcp_tool_call.rs

[codex] Remove string-keyed MCP tool maps (#21454)

2026-05-07 10:16:10 -07:00

mcp_tool_exposure_test.rs

[codex] Remove string-keyed MCP tool maps (#21454)

2026-05-07 10:16:10 -07:00

mcp_tool_exposure.rs

[codex] Remove string-keyed MCP tool maps (#21454)

2026-05-07 10:16:10 -07:00

mcp.rs

feat: make built-in MCPs first-class runtime servers (#21356)

2026-05-07 10:36:32 +02:00

memory_usage.rs

Remove ToolName display helper (#21465)

2026-05-08 12:17:48 -07:00

mention_syntax.rs

Extract codex-utils-plugins crate (#15746)

2026-03-25 11:05:35 -07:00

network_policy_decision_tests.rs

extract models manager and related ownership from core (#16508)

2026-04-02 23:00:02 -07:00

network_policy_decision.rs

refactor(proxy): clarify sandbox block messages (#17168)

2026-04-09 10:53:06 -07:00

network_proxy_loader_tests.rs

chore: clean up argument-comment lint and roll out all-target CI on macOS (#16054)

2026-03-27 19:00:44 -07:00

network_proxy_loader.rs

inline hostname resolution for remote sandbox config (#19739)

2026-04-27 03:18:57 +00:00

original_image_detail.rs

Remove js_repl feature (#19410)

2026-04-24 17:49:29 -07:00

otel_init.rs

codex-otel: add configurable trace metadata (#21556)

2026-05-07 16:06:57 -07:00

personality_migration_tests.rs

Revert state DB injection and agent graph store (#21481)

2026-05-06 22:48:29 -07:00

personality_migration.rs

Revert state DB injection and agent graph store (#21481)

2026-05-06 22:48:29 -07:00

prompt_debug.rs

extension: wire extension registries into sessions (#21737)

2026-05-11 11:38:18 +02:00

realtime_context_tests.rs

[codex-analytics] rework thread_source for thread analytics (#20949)

2026-05-06 02:12:31 +00:00

realtime_context.rs

[codex] Route live thread writes through ThreadStore (#18882)

2026-04-23 10:17:09 -07:00

realtime_conversation_tests.rs

Update realtime handoff transcript handling (#18597)

2026-04-20 14:04:09 -07:00

realtime_conversation.rs

Make realtime sideband startup async (#20715)

2026-05-04 22:28:14 +00:00

realtime_prompt.rs

[codex] Update realtime V2 VAD silence delay and 1.5 prompt (#18092)

2026-04-16 14:30:57 -07:00

review_format.rs

core: remove cross-crate re-exports from lib.rs (#16512)

2026-04-01 23:06:24 -07:00

review_prompts.rs

Spread AbsolutePathBuf (#17792)

2026-04-14 14:26:10 -07:00

rollout.rs

codex: use ThreadStore history for core review forks (#20577)

2026-05-05 15:25:19 -07:00

safety_tests.rs

permissions: remove cwd special path (#19841)

2026-04-27 13:41:27 -07:00

safety.rs

Make turn diff tracking operation backed (#21180)

2026-05-07 11:33:47 +02:00

sandbox_tags_tests.rs

permissions: make runtime config profile-backed (#19606)

2026-04-26 13:29:54 -07:00

sandbox_tags.rs

permissions: make runtime config profile-backed (#19606)

2026-04-26 13:29:54 -07:00

session_prefix.rs

Organize context fragments (#18794)

2026-04-20 22:39:17 -07:00

session_rollout_init_error.rs

remove temporary ownership re-exports (#16626)

2026-04-03 00:33:34 -07:00

session_startup_prewarm.rs

2- Use string service tiers in session protocol (#20971)

2026-05-06 18:00:21 +03:00

shell_detect.rs

chore: rename codex-command to codex-shell-command (#11378)

2026-02-10 17:03:46 -08:00

shell_snapshot_tests.rs

state: pass state db handles through consumers (#20561)

2026-05-04 11:46:03 -07:00

shell_snapshot.rs

state: pass state db handles through consumers (#20561)

2026-05-04 11:46:03 -07:00

shell_tests.rs

chore: clean up argument-comment lint and roll out all-target CI on macOS (#16054)

2026-03-27 19:00:44 -07:00

shell.rs

fix: add shell fallback paths for pwsh/powershell that work on GitHub Actions Windows runners (#16617)

2026-04-02 13:47:10 -07:00

skills.rs

Add plugin ID to skill analytics (#20923)

2026-05-04 20:36:29 -07:00

spawn.rs

Spread AbsolutePathBuf (#17792)

2026-04-14 14:26:10 -07:00

state_db_bridge.rs

state: pass state db handles through consumers (#20561)

2026-05-04 11:46:03 -07:00

stream_events_utils_tests.rs

Delete unused ResponseItem::Message.end_turn (#19605)

2026-04-26 17:18:09 -07:00

stream_events_utils.rs

Remove ToolName display helper (#21465)

2026-05-08 12:17:48 -07:00

test_support.rs

Revert state DB injection and agent graph store (#21481)

2026-05-06 22:48:29 -07:00

thread_manager_tests.rs

extension: wire extension registries into sessions (#21737)

2026-05-11 11:38:18 +02:00

thread_manager.rs

extension: wire extension registries into sessions (#21737)

2026-05-11 11:38:18 +02:00

thread_rollout_truncation_tests.rs

Delete unused ResponseItem::Message.end_turn (#19605)

2026-04-26 17:18:09 -07:00

thread_rollout_truncation.rs

[codex-analytics] feature plumbing and emittance (#16640)

2026-04-13 23:11:49 -07:00

turn_diff_tracker_tests.rs

fix: preserve exact turn diffs after partial apply_patch failures (#21518)

2026-05-07 18:05:45 +02:00

turn_diff_tracker.rs

fix: preserve exact turn diffs after partial apply_patch failures (#21518)

2026-05-07 18:05:45 +02:00

turn_metadata_tests.rs

feat: include thread ID in MCP turn metadata (#21329)

2026-05-06 11:36:15 +02:00

turn_metadata.rs

feat: include thread ID in MCP turn metadata (#21329)

2026-05-06 11:36:15 +02:00

turn_timing_tests.rs

Add turn start timestamp to turn metadata (#19473)

2026-04-28 16:36:59 +00:00

turn_timing.rs

[codex-analytics] add item lifecycle timing (#20514)

2026-05-04 22:33:20 +00:00

unavailable_tool.rs

Remove ToolName display helper (#21465)

2026-05-08 12:17:48 -07:00

user_shell_command_tests.rs

Organize context fragments (#18794)

2026-04-20 22:39:17 -07:00

user_shell_command.rs

Organize context fragments (#18794)

2026-04-20 22:39:17 -07:00

util_tests.rs

Update models.json (#18586)

2026-04-20 10:27:01 -07:00

util.rs

remove temporary ownership re-exports (#16626)

2026-04-03 00:33:34 -07:00

web_search.rs

add missing fields to WebSearchAction and update app-server types (#10276)

2026-01-30 16:37:56 -08:00

windows_sandbox_read_grants_tests.rs

core: remove cross-crate re-exports from lib.rs (#16512)

2026-04-01 23:06:24 -07:00

windows_sandbox_read_grants.rs

core: remove cross-crate re-exports from lib.rs (#16512)

2026-04-01 23:06:24 -07:00

windows_sandbox_tests.rs

Configure multi_agent_v2 spawn agent hints (#17071)

2026-04-08 08:42:18 -07:00

windows_sandbox.rs

Configure multi_agent_v2 spawn agent hints (#17071)

2026-04-08 08:42:18 -07:00