Commit Graph
1145 Commits
Author SHA1 Message Date
Adam Perry @ OpenAI 63002bdb26 Extract persisted history types into a dedicated crate (#37871)
## What changed

- Add `codex-history` for model-history and persisted-rollout domain types, including `RolloutItem`, `RolloutLine`, `CompactedItem`, and initial/resumed history state.
- Re-export the persisted types from `codex-rollout` and update consumers to use the new crate boundary instead of `codex-protocol`.
- Preserve existing rollout serialization, including legacy numeric compacted-window IDs.

## Testing

- Add `codex-history` tests for rollout JSON round trips, compacted-history compatibility, persisted history modes, and multi-agent version selection.

GitOrigin-RevId: 944daa9297ddd231d3aebbdcb05fff4adf8b4e1b
2026-08-10 19:26:52 +00:00
sayan-oai ee7815dad2 Rename environment config for turn scope (#37862)
## What changed

- Rename `EnvironmentConfig` to `TurnEnvironmentConfig` to make its scope explicit.
- Rename the corresponding `SessionConfiguration` accessor and update all call sites.

GitOrigin-RevId: f47d745566199e699bd8e7c671c313eaf3bba79e
2026-08-10 18:03:46 +00:00
Dylan Hurd d06dc73290 Route intercepted exec approvals through shared review (#37851)
## What changed

- Send Unix `execve` approvals intercepted by the zsh fork through the shared approval pipeline, including permission hooks, Guardian review, user prompts, and telemetry.
- Resolve the active turn and its auto-review setting when an intercepted command needs approval, so commands sent to persistent terminals use the current turn's reviewer.
- Give each intercepted command a distinct approval ID and propagate an aborted approval as a turn abort.

## Testing

- Cover Guardian review for intercepted `unified_exec` commands and persistent terminals across turns.
- Verify repeated identical intercepted commands receive separate user approvals.

GitOrigin-RevId: e6cccf160637e4246aff4714c22f08c90b65306d
2026-08-10 17:24:51 +00:00
jif 97729885d4 Expose the session ID to shell commands (#37848)
## What changed

- Set `CODEX_SESSION_ID` to the shared root-session ID for shell tool calls,
  unified exec processes, and user shell commands.
- Preserve `CODEX_SESSION_ID` when restoring a shell snapshot, alongside the
  existing runtime-owned environment variables.

GitOrigin-RevId: 5228867d25507105c7edf11607a8298958c3b917
2026-08-10 17:02:19 +00:00
Charlie Marsh beeba1d2fc Share model-visible tool specs across prompts (#37807)
## Why

Building a prompt cloned every model-visible `ToolSpec`, even though the tool
set is immutable for the lifetime of its router.

## What changed

- Store model-visible tool specs as an `Arc<[ToolSpec]>` in `ToolRouter` and
  `Prompt` so prompt construction only clones the shared pointer.
- Keep separately built routers on distinct shared slices so refreshed tool
  sets remain independent.

## Testing

Extend router tests to verify allocation sharing within a router, allocation
separation across rebuilt routers, and the existing deferred-tool filtering.

GitOrigin-RevId: 1e04cd7c4d1b3fa0b494c5c51670232d38ba8ebb
2026-08-10 10:50:49 +00:00
Charlie Marsh c9c6c0daa9 Add a feature flag to preserve apply_patch line endings (#37758)
## What changed

- Add the `apply_patch_preserve_line_endings` feature, disabled by default, to
  preserve CRLF, CR, and mixed line endings when `apply_patch` updates files.
- Apply the feature consistently to built-in patch handling and patches invoked
  through shell, user-shell, unified-exec, and app-server command execution.
- Keep the active feature configuration authoritative over inherited, shell
  snapshot, and client-provided environment values.

## Testing

- Cover line-ending behavior with the feature enabled and disabled for custom
  tool calls, shell heredocs, command execution, and the `apply_patch` CLI.

GitOrigin-RevId: 531a7c66761959c650270559f57941929f03e6c4
2026-08-10 01:44:30 +00:00
sayan-oai 420accf199 Use the step context for command approval prefix rules (#37641)
## What changed

Read `allow_prefix_rules` from the turn attached to the active step context
when selecting the exec policy and constructing unified exec approval requests.

GitOrigin-RevId: d2d26e88b9cefa28bb2df5b9f71aa7b30b7ea331
2026-08-09 00:29:14 +00:00
sayan-oai dd43a9967f Use step environments for Guardian approval reviews (#37618)
## Why

Deferred environments can become ready after a turn starts. Guardian approval
reviews must use the environment selected for the current step so their working
directory and permission context are not taken from the stale turn snapshot.

## What changed

- Carry `StepContext` through tool execution and approval handling.
- Add a Guardian review context that pairs the parent turn with the current
  step's environment snapshot.
- Use that snapshot for Guardian prompts, reviewer sessions, and delegated
  reviewer threads while retaining turn-level fallbacks for callers without a
  step context.

## Testing

Added an integration test that makes a deferred remote environment ready before
an escalated command and verifies Guardian receives that environment's working
directory and denied-read policy instead of the initial local environment's.

GitOrigin-RevId: 779096bc7d0e0c7453a009c02141029714b9b407
2026-08-08 20:07:27 +00:00
cooper-oai c4513cb982 Prevent launch context from reaching child processes (#37607)
## Why

Model-reachable child processes should not inherit Codex launch context.

## What changed

- Treat `OPENAI_FEDERATION_RULE_ID` and `OPENAI_IDENTITY_TOKEN_FILE` as non-inheritable environment variables, with case-insensitive matching.
- Remove them after shell environment policy overrides and before spawning commands across execution, MCP, hooks, Git helpers, and remote helper processes.

## Testing

- Cover inherited and explicitly configured variants, including mixed-case names.
- Verify the variables are absent from real child environments and app-server command and process execution.

GitOrigin-RevId: 2535527893985fef0995617f4c5b2462bea7c136
2026-08-08 16:58:26 +00:00
Abhinav 6f647caa9b Support asynchronous command hooks (#37533)
## Why

Hook configurations can mark command handlers as asynchronous, but Codex previously skipped those handlers outside `SessionEnd`.

## What changed

- Run asynchronous command hooks in the background with a per-session concurrency limit, while keeping `SessionEnd` hooks synchronous.
- Prevent asynchronous hooks from blocking, stopping, rewriting, or otherwise controlling the operation that launched them.
- Deliver warnings and additional context at safe turn boundaries: inject results into an active turn after sampling, or buffer them ahead of the next user prompt when the session is idle.
- Preserve in-flight hooks across configuration reloads, scope spilled output to the thread, and abort outstanding work during session shutdown.

## Testing

Add unit and integration coverage for background scheduling, concurrency, output parsing, active and idle result delivery, configuration reloads, and shutdown cleanup.

GitOrigin-RevId: 8094552e4afe7b47b09a61bb575bb20f4e491d8d
2026-08-08 04:24:12 +00:00
andrewgu-oai e734a1a5c1 Ignore reusable command approvals for cyber models (#37516)
## What changed

- Filter saved `allow` prefix rules from the execution policy for cyber-specialized models and models listed in `auto_review.ignore_rules`, while preserving prompt, forbidden, network, and host-executable policy entries.
- Apply the filtered policy consistently to permission instructions, shell commands, unified exec, and the zsh-fork backend.
- Limit cyber-model command approvals to one-time decisions without proposing reusable policy amendments. General models continue to honor saved prefixes, including after switching models within a thread.

## Testing

- Add policy-level coverage for filtering allow-prefix rules while retaining restrictive and network rules.
- Add end-to-end coverage across shell and unified exec paths, zsh-fork execution, user and automatic review, and model switching.

GitOrigin-RevId: f3bd3872424291b12354ec415f33986ec369a368
2026-08-08 00:59:39 +00:00
rka-oai 8e4b10446e Remove the legacy code-mode tool metadata inventory (#37500)
## What changed

- Stop adding `code_mode_tool_names` to Responses Lite turn metadata.
- Keep `code_mode_tool_names` reserved so client metadata cannot reintroduce the removed, unbounded inventory.
- Use the opt-in `tool_namespaces_info` metadata, including each function's `code_mode_name`, when authoritative tool details are requested.

## Testing

- Update Responses Lite and code-mode tests to verify the opt-in namespace metadata and normalized code-mode names.

GitOrigin-RevId: cbfb9af984de71c7cc175d0cd47ae799a5275309
2026-08-07 20:26:08 +00:00
rka-oai 2b1811e562 Include tool namespace inventory in turn metadata (#37492)
## What changed

- Add opt-in `tool_namespaces_info` metadata for Responses Lite turns when
  `tool_registry.turn_metadata_includes_tool_info` is enabled.
- Describe each model-visible function's namespace, direct and deferred
  exposure, Code Mode name, and harness or MCP ownership.
- Reject visible namespaces shared by different owners when strict collision
  checking and tool inventory metadata are enabled.
- Keep tool inventories out of compatibility headers and metadata sent to MCP
  servers.

## Testing

- Cover opt-in behavior, MCP exposure details, namespace ownership collisions,
  reserved metadata handling, and Responses Lite request serialization.

GitOrigin-RevId: 6a6965d2ba98e29444fdfbb4ccec6aafcf0391e6
2026-08-07 19:32:15 +00:00
pakrym-oai 509565820f Interrupt active code-mode cells with their turn (#37483)
## Why

Interrupting a turn should also stop code-mode work that the turn left running.

## What changed

- Add the disabled-by-default `code_mode_interrupt` feature.
- When an interrupted turn has the feature enabled, terminate all active code-mode cells and propagate cancellation through nested tool calls.
- Keep the reusable code-mode session alive so its stored state remains available to later turns.

## Testing

Add an integration test that interrupts a long-running nested tool, verifies that active and background cells are removed, and confirms that session state is preserved.

GitOrigin-RevId: 2b8634776a89b636318a39b9e9ad2eefc07cead8
2026-08-07 18:22:42 +00:00
iceweasel-oai 511262b984 Delegate remote process sandboxing to the executor (#37480)
## What changed

- Preserve executor-native working directories, workspace roots, and permission profiles when preparing remote `exec_command` requests instead of resolving them through the host platform.
- Send sandbox intent to the remote executor, including a restricted-token default for Windows executors and `-NoProfile` for elevated PowerShell commands.
- Reject intercepted cross-platform `apply_patch` commands when filesystem writes are restricted, while continuing to allow them for unsandboxed or full-disk-write profiles.

## Testing

Extended remote unified-exec coverage for foreign Windows workspaces, elevated PowerShell, sandbox metadata, and restricted and unrestricted intercepted patches.

GitOrigin-RevId: 9fd328879fa2c15594fd58e40b6b32e8ba0847e7
2026-08-07 18:05:17 +00:00
jif 964a227d8c Preserve base instruction provenance across sessions (#37446)
## Why

Persisted base instructions need to retain whether they were explicitly customized or generated from a model template. Without that distinction, forks and config-lock replays can treat model-generated instructions as custom and keep the wrong template after a model or personality change.

## What changed

- Record custom or model provenance with base instructions in rollouts and config locks.
- Preserve custom instructions across model changes, while allowing model-generated instructions to follow the selected model and personality.
- Keep legacy rollouts without provenance compatible and propagate provenance through forks and spawned agents.

## Testing

- Cover provenance serialization and legacy rollout decoding.
- Cover role personality changes, forked model changes, config-lock replay, and multi-agent configuration.

GitOrigin-RevId: 801c26559835f558027ce5112700c0cd31fcb21b
2026-08-07 14:08:13 +00:00
felixxia-oai e58d9ef447 Unify plugin skill loading with the host skill service (#37444)
## What changed

- Inject the host skill loader into `PluginsManager` so plugin discovery and agent turns use the same loading and product-policy behavior.
- Share plugin skill snapshots across those paths, preserving a consistent view of skills across workspaces.
- Apply migrated-command precedence after product filtering, allowing an eligible migrated command to replace a filtered native skill with the same name.

## Testing

- Add coverage for product-restricted plugin skills, native-versus-migrated command precedence, and the skills exposed to agent turns.

GitOrigin-RevId: f5ef0d0766ebeeb30d73ffaf044d003c2906ea4d
2026-08-07 13:54:24 +00:00
Adam Perry @ OpenAI 4d7e3e90d9 Clamp short wait_agent timeouts to the configured minimum (#37357)
## What changed

- Clamp `wait_agent` timeout requests below `min_wait_timeout_ms` instead of rejecting them.
- Include the timeout adjustment in the wait result message and document it in the tool output schema.

## Testing

- Update the handler test to verify the configured minimum is honored and the adjustment is reported.
- Update the schema test for the expanded result message.

GitOrigin-RevId: 2b4a770ed6d1e9c45b351b479f5ae50c9a3d5388
2026-08-07 02:34:46 +00:00
pakrym-oai d0c8f422ea Configure the default code-mode exec yield timeout (#37352)
## What changed

- Add `features.code_mode.default_exec_yield_time_ms`, defaulting to 30 seconds.
- Apply the configured default to code-mode `exec` calls that omit `yield_time_ms` and reflect it in the tool description.
- Remove the `code_mode_buffered_exec` compatibility feature flag.

## Testing

- Cover config loading and tool descriptions for both default and custom yield timeouts.

GitOrigin-RevId: 581ffc737b04721c7b5fc81260b1da1cf7d87523
2026-08-07 01:41:54 +00:00
jif 57f42a8113 Avoid cloning immutable metadata on tool search cache hits (#37279)
## Why

Checking the tool search cache currently rebuilds search metadata for every deferred tool, even when an immutable tool runtime has not changed.

## What changed

- Track immutable deferred tools by runtime identity and dynamic tools by their current search metadata.
- Rebuild the search handler when the deferred registry, dynamic metadata, or source-listing mode changes.
- Keep only weak references to immutable runtimes in the cache.

## Testing

Added unit coverage for immutable handler reuse, registry and exposure changes, source-listing changes, and refreshed dynamic tool metadata.

GitOrigin-RevId: 024c6f4498d431e7bbfaecde420f77fcf8c0dba7
2026-08-06 14:11:35 +00:00
jif e1831db7c3 Reuse MCP handlers across sampling steps (#37273)
## Why

MCP tool handlers and their Code Mode definitions are immutable for the lifetime
of a stable MCP binding, so rebuilding them for every sampling step repeats schema
construction unnecessarily.

## What changed

- Cache MCP handlers per session and reuse them while the same binding remains
  active, invalidating the cache when the binding changes.
- Lazily cache schema-free Code Mode definitions on MCP handlers while continuing
  to rebuild definitions for step-local tools whose specifications may change.
- Reapply current app enablement, tool policy, exposure, and plugin schema budgets
  whenever cached handlers are registered for a step.

## Testing

- Cover handler and Code Mode definition reuse, binding invalidation, changing app
  policy, stable continuation schemas, and step-local custom tool updates.

GitOrigin-RevId: 69c5011fb97c08580ac476e7fb2286b1e92e0359
2026-08-06 13:37:07 +00:00
jif 82b17bc724 Allow agent roles on full-history forks (#37252)
## Why

Multi-agent v2 rejected `agent_type` when `fork_turns` was `all`, forcing callers to choose between a configured role and the parent's full conversation context.

## What changed

- Apply an explicitly selected role to full-history forks, including its model and developer instructions.
- Preserve the parent's developer instructions when the selected role does not define its own.
- Keep the inherited conversation history and model-switch context intact.
- Update the `spawn_agent` schema to clarify that roles apply independently of history inheritance.

## Testing

- Cover full-history role overrides and developer-instruction precedence in multi-agent v2 tests.

GitOrigin-RevId: cd4f35e31d75f56289760f245bff38dd0233abe6
2026-08-06 10:27:33 +00:00
Curtis 'Fjord' Hawthorne 0a0ebb8535 Add a unified image budget (#37206)
## What changed

- Add the gated `unified_image_budget` feature for models that support original image detail or Responses Lite.
- Apply one 6,000-pixel, 10,000-patch preprocessing limit regardless of legacy image detail hints.
- Hide detail controls from `view_image` and code mode while continuing to accept existing hints for compatibility.
- Preserve detail-based resizing and tool contracts for unsupported models.

## Testing

- Cover unified resizing limits, Responses Lite, legacy detail hints, unsupported models, and `view_image` integration in code mode.

GitOrigin-RevId: cb07bff1669a96599fdfd076b3d9ec80f2b6fff7
2026-08-06 03:15:08 +00:00
Adam Perry @ OpenAI 92b83e226d Track multi-agent usage hints in world state (#37189)
## Why

Resumed sessions need the current multi-agent usage instructions when configuration changes or the saved history predates usage-hint tracking.

## What changed

- Store multi-agent usage hints in world state and refresh changed or previously untracked hints before re-emitting the active multi-agent mode.
- Present `wait_agent` polling guidance in the overridable developer instructions only when the tool is enabled, instead of embedding it in the tool schema.

## Testing

- Cover cold resumes, changes to `wait_agent` availability, custom usage hints, message ordering, and tool-schema behavior.

GitOrigin-RevId: a9c6798d5dc2dedc8dadc49e0a80ff50a757ab83
2026-08-06 00:22:06 +00:00
tongzhou wang 98da2c4499 Reserve the tool_search namespace for the search tool (#37188)
## What changed

- Remove namespace tools named `tool_search` before registering the built-in
  search tool, preventing them from sharing its model-visible surface.
- Record removed namespace tools as collisions so strict collision handling
  reports their qualified tool names.

## Testing

- Cover direct and deferred MCP tools that use the reserved namespace.
- Cover strict collision errors for both direct and deferred conflicts.

GitOrigin-RevId: 28fff918b4358a67b167467d92728fa545b951d7
2026-08-06 00:16:57 +00:00
jimxie-oai 928bda82cf Preserve image transparency metadata in app-server items (#37178)
## What changed

- Add `transparentBackground` to image-generation items and legacy completion events.
- Map Images API backgrounds to `true` for transparent, `false` for opaque, and `null` for automatic or unavailable values.
- Preserve the metadata in live notifications and persisted or resumed thread history while continuing to deserialize older items that omit it.

## Testing

- Cover transparent, opaque, and automatic backgrounds, including image edits and thread read/resume flows.
- Verify the stable JSON shape and optional TypeScript field.

GitOrigin-RevId: 08527f37f23431df45fab1ed34eb82b2139b49d7
2026-08-05 23:42:29 +00:00
Dylan Hurd 778b869829 Centralize tool approval handling in Session (#37128)
## What changed

- Move permission hooks, reviewer routing, approval caching, and user approval requests into the session-level approval flow.
- Have shell, unified exec, and apply-patch runtimes describe requests as `ApprovalAction` values, which now provide their hook payloads and cache keys.
- Pass a shared `ApprovalContext` through initial and retry approvals so telemetry and approval reasons are handled consistently.

GitOrigin-RevId: 5bfa61daa97e4427f3a648f6cf9d1c27f1551942
2026-08-05 17:48:45 +00:00
jif fcc4ca552f Preserve longer MCP source descriptions in tool search (#37066)
## Why

The 4 KiB aggregate limit can truncate source metadata before it reaches the
model through the tool search description.

## What changed

Raise the aggregate source-description limit to 512 KiB while retaining the
existing character-boundary-safe truncation and bounded output.

## Testing

Expand unit and RMCP round-trip coverage to verify that substantially longer
Unicode descriptions remain complete within the new limit.

GitOrigin-RevId: 002760cdd2e8741e531b45f75f8624ef84334d24
2026-08-05 09:39:05 +00:00
Won Park 757c151a0e Add safer TUI defaults for cyber models (#37055)
## What changed

- Propagate the optional `modelSpecialty` field through the model catalog and
  app-server `model/list` response.
- When a newly selected model has the `cyber` specialty, default an active TUI
  thread to workspace-write permissions with on-request approval. Use auto
  review when available, otherwise keep the user as reviewer, while respecting
  configured permission requirements.
- Show a notice when auto review is applied and strengthen the full-access
  warning for cyber models.
- Preserve explicitly selected permissions when only reasoning settings change.

## Testing

- Cover auto-review defaults, requirement-aware fallback behavior, permission
  preservation across reasoning changes, and the new warning notice.

GitOrigin-RevId: 8f25b0082d3051d9ff91f7355921236d0f82f758
2026-08-05 08:04:15 +00:00
rka-oai 3171166881 Reject conflicting namespace descriptions in strict tool mode (#37053)
## What changed

- Treat different non-empty descriptions for the same tool namespace as a
  collision when `error_on_tool_collisions` is enabled, including deferred
  tools.
- Continue allowing multiple tools to share a namespace when their descriptions
  match or are empty.
- Preserve the first non-empty namespace description when strict collision
  checking is disabled.

## Testing

- Cover strict collisions across direct and deferred tools, valid shared
  namespaces, relaxed description merging, and failure before model sampling.

GitOrigin-RevId: 788a2ae0eccb64c97995af7a208d9abb79edf2d0
2026-08-05 08:01:47 +00:00
sayan-oai bac3ef1d8e Use turn environment permissions for tool execution (#37038)
## What changed

- Derive sandboxing, approval, and permission-profile environment variables from the selected turn environment for shell, unified exec, `apply_patch`, image viewing, and network approval flows.
- Materialize project-root permissions against every workspace root in that environment, allowing `apply_patch` to write to secondary workspace roots while preserving protected paths.
- Carry the selected environment's permission profile into spawned or resumed agents and Guardian permission context.

## Testing

- Add an integration test covering `apply_patch` writes in a secondary workspace root.
- Extend tool, Guardian, and multi-agent tests to distinguish environment permissions from thread-level permissions.

GitOrigin-RevId: e22da8974bcd3de1e724978b1a7790307a51e4bc
2026-08-05 06:38:57 +00:00
sayan-oai 1fe6be9719 Align registry tests with canonical tool names (#37035)
## What changed

- Expect tool collisions to report names canonicalized into the default
  `functions` namespace.
- Remove the obsolete test that allowed an explicitly namespaced
  `functions.shell_command` external tool.

GitOrigin-RevId: 9a517f49c027ffe5b811f6668dca91246f4925a6
2026-08-05 06:07:53 +00:00
sayan-oai 30d99232f4 Apply permission profile updates to future turn environments (#37031)
## What changed

- Include the current permission profile in each turn environment's configuration.
- Refresh retained environment configurations when session settings change, while preserving the configuration already captured by active turns.
- Keep environment connection and shell snapshot resolution reusable across configuration updates.

## Testing

- Cover permission profile updates applied through both session settings updates and per-turn updates.
- Verify that pending environment resolutions use the configuration captured by their turn and that configuration-only updates do not retry failed environments.

GitOrigin-RevId: 923694f0729b87a118d7635ade3a8ca9bcd4e7b1
2026-08-05 05:30:51 +00:00
jacobzhou-oai 56b82e676c Enforce Agent Plugin runtime boundaries (#37027)
## What changed

- Track Agent Plugin manifests through plugin, skill, and MCP loading so their capabilities use format-specific behavior without changing legacy plugins.
- Discover only direct-child skills, exclude app and hook capabilities, isolate MCP data, and reject MCP configuration files that are non-regular or resolve outside the plugin root.
- Bound model-visible skill instructions, plugin instructions, MCP descriptions, schemas, individual tools, and the aggregate Agent Plugin MCP tool set.
- Stop MCP and OAuth redirects when Agent Plugins send configured or authorization headers, while retaining existing redirect behavior for legacy MCP servers.

## Testing

- Add coverage for capability filtering, skill discovery, isolated MCP data and reserved-path expansion, unsafe MCP configuration files, context limits, and redirect handling.

GitOrigin-RevId: c9af66b051269f3226628ca280a58d32c808c38f
2026-08-05 04:54:33 +00:00
rka-oai f21dc46388 Canonicalize default tools under the functions namespace (#37022)
## What changed

- Group top-level function and custom tool definitions into a single `functions` namespace for Responses Lite providers that support namespaced tools.
- Normalize missing, empty, and explicit `functions` namespaces to the same tool identity across registration, routing, lifecycle hooks, configuration, and tool search.
- Keep default tool names unprefixed in code mode, display output, and dispatch traces while preserving explicit non-default namespaces.

## Testing

- Add coverage for Responses Lite serialization, tool search results, namespace normalization and collision handling, routing, lifecycle events, and code-mode namespace policies.

GitOrigin-RevId: d48414005b5d22d39b11a198e19c47814d3a19f2
2026-08-05 04:07:59 +00:00
rka-oai 1e489adad0 Enforce strict tool name collision errors (#37020)
## What changed

- Track the first duplicate effective tool name while assembling the tool registry, including collisions introduced by external tools, code mode, and tool search.
- When `[features.tool_registry].error_on_tool_collisions` is enabled, fail the turn with a `duplicate tool: <namespace>.<name>` error before sending a model request.
- Continue allowing identical tool names in different namespaces, and preserve the existing collision behavior when strict checking is disabled.

## Testing

- Cover registry collision tracking, strict tool planning across tool sources, namespace isolation, failure before sampling, and pre-sampling compaction.

GitOrigin-RevId: 4cf83538fb513cec0b9c8b567780caaaadd3243d
2026-08-05 04:00:50 +00:00
andrewgu-oai 11e390bb10 Remove legacy collaboration mode variants (#36990)
## What changed

- Remove the hidden `PairProgramming` and `Execute` variants from `ModeKind`.
- Delete their unused prompt templates and simplify mode handling to cover only `Default` and `Plan`.

GitOrigin-RevId: 8d24e3f562a09e3071b2decae275d2b3871a1e9c
2026-08-04 22:58:02 +00:00
Adam Perry @ OpenAI 78f00743f9 Allow disabling the built-in image viewer (#36966)
## What changed

- Add the stable, default-enabled `features.view_image` flag.
- Omit the native `view_image` tool when the flag is disabled, including for fresh-context subagents and guardian reviewer turns.
- Keep unrelated execution and MCP tools available when the viewer is disabled.

## Testing

- Add app-server coverage for disabled viewer inheritance in fresh-context subagents and guardian reviewer turns.

GitOrigin-RevId: 42b1311010fbde9f064129b17b425820861c6d91
2026-08-04 20:10:31 +00:00
Dylan Hurd 2b1357c27c Include policy approval reasons in Guardian reviews (#36939)
## Why

Guardian reviews received sandbox retry reasons, but not the execution-policy reason that triggered an initial approval request.

## What changed

- Propagate approval and retry reasons through the tool approval flow and include the applicable reason in the Guardian prompt.
- Prefer a sandbox retry reason when both are available.
- Truncate approval context to 512 tokens while preserving the start and end of the reason.

## Testing

Added prompt and integration coverage for policy reasons, retry precedence, and truncation.

GitOrigin-RevId: 5b0f4e1c40b792b031bc8f7b31685a61d12b8d92
2026-08-04 18:01:04 +00:00
Celia Chen 3ca9f375aa Enable cached web search for Amazon Bedrock (#36938)
## Why

Amazon Bedrock supports hosted text web search, but it rejects the
`search_content_types` field used for multimodal search and does not support
external live or indexed web access.

## What changed

- Advertise hosted web search for Amazon Bedrock while marking external web
  access as unsupported.
- Resolve unsupported live and indexed modes to cached search, or disable the
  tool when cached search is prohibited by managed requirements.
- Normalize built-in and configured Bedrock model catalogs to text-only web
  search, and retain the runtime provider in session configuration so turn
  setup can apply its capabilities.

## Testing

- Cover cached fallback, managed-mode restrictions, text-only tool payloads,
  provider capabilities, and catalog normalization.

GitOrigin-RevId: 310473849257401654388a4ebb42920e03aa3228
2026-08-04 17:55:28 +00:00
jif 90314a9207 Read turn permissions from the current configuration (#36930)
## What changed

- Remove the cached permission profile from `TurnContext` and derive the
  effective profile, filesystem policy, network policy, and legacy sandbox
  policy from `config.permissions`.
- Update sandbox consumers and tests to use the current turn configuration.
- Verify that role-based agent spawning reapplies runtime permissions and that
  cold-resumed agents retain their disabled permission profile.

GitOrigin-RevId: d67c5d2bcbe6dc76f15b56defb485b155fd1f138
2026-08-04 17:13:13 +00:00
jif c8e255e7f8 Centralize app enabled-state evaluation (#36916)
## What changed

- Add `AppToolPolicyEvaluator::apply_app_enabled_state` and use it when
  presenting app lists, building plugin context, and deciding whether app
  instructions are available.
- Preserve each app's source state unless local or managed configuration
  explicitly overrides it.
- Keep connector discovery and post-install refresh checks based on raw
  accessibility rather than configured enablement.

## Testing

- Cover default enablement, per-app overrides, managed disablement, and
  preservation of unconfigured source state.

GitOrigin-RevId: f1a62d55e7cc48b37113848e3baa0d69d8d9c8a8
2026-08-04 15:49:49 +00:00
jif 6a828ca26f Read approval policy from the current turn configuration (#36912)
## Why

Thread settings can update the approval policy after a turn context is created.
Keeping a separate copy on `TurnContext` could leave tool approval checks using
the previous policy.

## What changed

- Remove the duplicated approval-policy field from `TurnContext`.
- Resolve the policy through the turn's current configuration everywhere it is
  needed, including tool execution, Guardian routing, MCP handling, and
  permission requests.

## Testing

- Update the shell escalation test to apply a thread-level policy override and
  verify that the next turn rejects an escalation using the updated policy.

GitOrigin-RevId: e5966ba08f179d53fd76871ad904762958c0f5ea
2026-08-04 15:32:31 +00:00
Martin Au-Yeung ee46c5ba0e Negotiate MCP extensions per app-server session (#36910)
## Why

App-server clients need to advertise structured MCP extension settings, including supported MCP App UI MIME types, rather than only opting into OpenAI form elicitation with a boolean.

## What changed

- Add an `extensions` map to initialize capabilities and preserve `mcpServerOpenaiFormElicitation` as a legacy alias for `openai/form`.
- Capture the declared extension profile when a thread is started, resumed, or forked, propagate it to subagents, and advertise it to downstream MCP servers during initialization.
- Keep the profile stable for the lifetime of the loaded session instead of changing it on later turns or direct tool calls.

## Testing

- Cover extension conversion, downstream MCP initialization, session isolation, legacy form support, and subagent inheritance.

GitOrigin-RevId: fbcedbb74ce788e574b0f884a4c45c4cedb9de54
2026-08-04 15:23:43 +00:00
jif 18f03c1eb7 Register app tools independently of the connector list (#36900)
## What changed

- Gate Codex Apps MCP tool registration on whether apps are enabled instead of requiring each tool's connector to appear in the accessible connector list.
- Continue enforcing model-visibility and app-tool policy checks, including the requirement for connector metadata.
- Use the merged connector catalog for tool-suggestion discovery.

## Testing

- Cover app-tool registration from catalog metadata, including synthetic links, source ordering, and the apps-disabled case.

GitOrigin-RevId: eef4eb03da9738d3a165dd6cfda0d3c1844d51e7
2026-08-04 14:19:13 +00:00
jif 6d4d9442c7 Support leaf models in multi-agent v2 (#36892)
## What changed

- Allow multi-agent v2 parents to spawn any visible model that has not explicitly disabled multi-agent support.
- Expose collaboration tools to child agents only when their selected model supports multi-agent v2, keeping legacy models as leaf workers.
- Preserve a worker's selected model when reloading it into residency.
- Propagate multi-agent capability metadata to Amazon Bedrock model entries so delegation is gated consistently.

## Testing

- Cover model selection, leaf-worker tool visibility, Bedrock capability handling, and model preservation after reload.

GitOrigin-RevId: 1457adb1a09806d1f0621311d5a42a6815b9dd4e
2026-08-04 12:36:14 +00:00
jif fd1e4d7a6d Preserve complete MCP namespace descriptions (#36882)
## What changed

- Keep complete MCP namespace descriptions in tool-search source metadata.
- Raise the namespace tool-spec description limit from 1,000 bytes to 512 KiB,
  truncating at a UTF-8 character boundary only when the new limit is exceeded.

## Testing

- Cover descriptions beyond the former limit and multibyte truncation at 512 KiB.
- Update SSE and stdio MCP tests to verify that complete server instructions are
  preserved without hiding tools.

GitOrigin-RevId: 000bfcafb3df348065ae451685bfbf978a0e3248
2026-08-04 11:14:06 +00:00
rka-oai d4fb78bfc5 Support custom tools in namespaces (#36857)
## What changed

- Allow namespace tool specs to contain custom freeform tools alongside function tools.
- Include namespaced custom tools in deferred tool search and expose them to code mode with names such as `editor__apply_patch`.
- Route custom payloads to matching extension tools while preserving function-only payload validation.

## Testing

- Add serialization, tool search, code-mode definition, and end-to-end dispatch coverage for namespaced custom tools.

GitOrigin-RevId: be64d35f6ae54685c5a9fcf45a732320742ea7e5
2026-08-04 08:28:43 +00:00
rka-oai 12288240b4 Support deferred loading for freeform tools (#36856)
## What changed

- Add optional `defer_loading` support to freeform Responses API tool definitions.
- Omit the field when it is unset so existing eager tool definitions retain their wire shape.

## Testing

- Verify legacy freeform tool deserialization and eager and deferred serialization shapes.

GitOrigin-RevId: 50e8658a54ac5b6ae0c1dbfe65f7bb62efef561d
2026-08-04 08:24:42 +00:00
Dylan Hurd b2dc8b3e4b Consolidate approval telemetry context (#36825)
## What changed

Add the tool name to `ApprovalCtx` and use the context's call ID and session
telemetry when recording approval decisions. This removes redundant `ToolCtx`
and telemetry parameters from `resolve_tool_approval`.

GitOrigin-RevId: 238a6f708f83993d64606bfaa74df344b0b46178
2026-08-04 02:29:43 +00:00