- Add `AlignOpenAIToolCallMessages` to reorder tool results immediately after the matching assistant tool calls while preserving original content and numeric precision.
- Prevent deferred message reordering and call ID guessing when tool outputs are incomplete, duplicate, or missing IDs.
- Normalize translated requests after applying summary configuration in Codex multi-agent execution.
Closes: #5925
Codex upstreams can silently serve a different model than the one requested
(HTTP 200, with response.model naming the substitute). The proxy kept no record
of it: nothing logged, nothing reported, only the pass-through response body.
- Add Record.ResponseModel to sdk/cliproxy/usage, aligned with the existing
ResponseServiceTier field, and emit it from the redis usage queue as the
optional response_model payload field alongside response_service_tier. Only
the record for the requested model carries it: additional-model records
(image generation tool usage) describe a side model the upstream response
never refers to, and would otherwise look like a substitution downstream.
- Add internal/runtime/executor/helps/response_model.go with
extractCodexResponseModelEvent (SSE frames and raw JSON, restricted to the events
that embed the authoritative response object, rejecting non-string and
oversized upstream model names) and IsCodexModelSubstituted (both sides
trimmed, lower-cased and stripped of thinking suffixes, dated aliases such as
gpt-5.6-terra-2026-05-13 accepted in either direction).
- UsageReporter records the served model on the event path and emits the WARN
when the attempt publishes its usage record, so no logging work happens
before the first event is forwarded. Repeats are throttled per
(auth id, requested model, served model) with a 10 minute window, because on
an affected credential every request is substituted and an unthrottled
warning would mirror the whole request volume into the logs. The credential
is labelled auth_index=<index> only: codex credential file names embed the
account e-mail, which must not be written to the logs at request rate.
Coverage, by entry point. The served model is observed on the HTTP streaming
path (both the bootstrap-buffered handshake and the streaming goroutine), the
HTTP non-streaming Execute loop, the websocket streaming and non-streaming
paths, and the two /responses-shaped image entry points. The remaining codex
entry points cannot report it and are therefore left alone: executeCompact
(/responses/compact answers with a compaction object that has no event type and
no response.model), the two direct image endpoints (/images/generations and
/images/edits answer in the Images API shape and stream image_generation.*
events), and CountTokens (counts locally with tiktoken, never reaching an
upstream).
TokenAccountingSchemaVersion is not bumped: it versions the token accounting
contract (token breakdown semantics), and this change only adds an optional
non-token field that leaves existing consumers and all token math untouched.
Note: response_model ships with the usage record and is the counting source;
the WARN is a throttled alerting signal and must not be used to count
substitutions.
Tests: table-driven unit tests for both helpers over real model ids, reporter
tests covering the published record, the single throttled warning, the absence
of account identifiers in it, concurrent observation and publishing under
-race, the throttle window and its entry bound, an executor-level guard for the
observeCodexTokenEvent wiring and the per-model records, plus a redisqueue
payload assertion for response_model. gofmt, go vet, go test -race on the
touched packages and go test ./... are clean.
- Normalize `service_tier` by trimming whitespace and matching case-insensitively.
- Map `fast` to `priority` and preserve `ultrafast` service tier.
- Strip unsupported service tier values and non-string types before forwarding.
Closes: #5924
- Detect Devin models by ID prefix, type, ownership, model registry metadata, or provider.
- Append `(Devin)` to model display names in Codex client responses when not already present.
- Set model `type` to `devin` in home Codex model formatting when served by the Devin provider.
- Restore client-defined `web_search` tool names from their alias across SSE, websocket, and buffered execution responses.
- Preserve namespaced tool calls when matching and restoring tool names.
Closes: #5923
- Strip `prompt_cache_breakpoint` from `output` arrays in function call outputs.
- Remove item-level `prompt_cache_breakpoint` fields from input items.
Closes: #5922
- Skip parsing the `Retry-After` header when a rate limit rejection is overage-only to prevent global credential cooldowns.
- Allow exponential backoff to handle model recovery while keeping shared subscription windows available for other models.
Closes: #5920
- Generalize forced hosted tool choice handling across both image generation and web search.
- Normalize forced web search tool choices to string mode and isolate the target tool in the tools list.
- Strip hosted web search from mixed `allowed_tools` definitions.
- Skip native `x_search` injection when a hosted tool is exclusively required.
Closes: #5916
- Add `claude.model-level-cooling` configuration to scope rate limit cooldowns to the requested model.
- Treat overage-only and spend cap rejections as model-scoped when shared subscription windows remain healthy.
- Propagate model-level cooling settings into streaming, token counting, and direct execution error classifiers.
Closes: #5915
- Match tool results against pending tool calls and downgrade unmatched results to user messages.
- Prevent downgraded orphaned tool results from consuming images intended for user turns.
- Unwrap protocol wrapper envelopes and extract structured text parts while preserving arbitrary business JSON.
- Provide a placeholder for empty or whitespace-only tool results.
Closes: #5911
- Track and aggregate tool calls by call ID instead of slot index in streaming and buffered execution.
- Support raw arguments from invalid JSON fields for custom tool calls.
- Parse usage field 4 as cache write tokens instead of adding to prompt tokens.
- Align client metadata with the default client name and drop deprecated tag 28.
Closes: #5910
- Return an error when the `$TOKEN$` placeholder cannot be resolved or the credential for `auth_index` is missing.
- Ensure `$TOKEN$` substitutions in headers and request payloads fail fast instead of proceeding with empty values.
- Support resolving and refreshing provider OAuth tokens for API calls.
Closes: #5838
- Add `SchedulerAcrossPriorities` capability to allow plugin schedulers to receive candidates across all priority tiers.
- Propagate scheduler priority preferences through the plugin host and RPC capabilities.
- Update auth manager selection logic to supply candidates across all priorities when the scheduler opts in.
Closes: #5894
- Replace the plugin configs subtree directly instead of merging to prevent zero values, booleans, and empty collections from being pruned as defaults.
- Ensure stale plugin keys and removed configuration entries are properly cleaned up while preserving comments.
Closes: #5907
- Extract and attach images from tool results to corresponding tool prompts by call ID.
- Preserve structured business JSON and raw objects in function result content.
- Prepend image headers to tool prompt content when images are present.
Closes: #5893
- Filter out `automation_update` tools and sanitize tool descriptions in Devin wire requests and logs.
- Strip additional Codex prompt directives from system messages.
- Support `children` field fallback when collecting namespace tools.
- Replace tool image placeholders with omission markers for compatibility with text-only upstream models.
- Strip synthetic image relay notices and image parts from user messages.
Closes: #5884
- Add reusable `SystemReminderText` helper to format directives within `<system-reminder>` envelopes.
- Wrap demoted mid-session system and developer messages in Gemini and Antigravity OpenAI request translators so upstream models treat them as system directives.
Closes: #5878
- Buffer content and tool call deltas while thinking is active so late-arriving thinking signatures can be attached before closing thinking blocks.
- Ensure pending actions and open steps are properly flushed and closed on stream completion or trailer errors.
Closes: #5873
- Extract reasoning text sequentially from `reasoning_content`, `reasoning`, and `reasoning_details`.
- Unify reasoning extraction across streaming deltas and non-streaming message conversions.
Closes: #5872
- Add incremental merging for Gemini `groundingMetadata` with chunk index remapping and query deduplication.
- Implement rune offset mapping across multipart messages for accurate `url_citation` annotations.
- Manage full streaming lifecycle for web search calls, emitting `searching`, `completed`, and output item done events.
- Stream incremental citation annotations via `response.output_text.annotation.added` events.
- Support `web_search_preview_2025_03_11` as a recognized web search tool type.
- Break stream scan loops immediately when upstream completion is reached.
- Skip scanner error handling and cancellation checks when `upstreamCompleted` is true.
Closes: #5866
Devin upstream occasionally encodes transient capacity failures using
Connect code permission_denied with message containing 'high demand'.
CPA's ParseDevinTrailerError currently maps every permission_denied to
HTTP 403. The auth cooldown manager interprets 403 as a 30-minute model
permission cooldown, keeping a recovered model locally unavailable.
This fix narrowly reclassifies the observed high-demand variant as
HTTP 429 (Too Many Requests), so it enters the quota/retry cooldown
path instead of the long permission denial path. Genuine
permission_denied errors (model access denied, plan entitlement denied,
etc.) remain HTTP 403.
Tests: 4 new cases in TestParseDevinTrailerError covering transient
high-demand (429), genuine permission error (403), resource_exhausted
unchanged (429), and case-insensitive matching. All existing tests
pass with no regressions.
- Track tool names by tool use ID from assistant tool call blocks.
- Populate the `name` field on converted tool result messages when a matching ID exists.
Closes: #5859
- Add bidirectional web search translation between OpenAI Responses API and Gemini/Antigravity
- Map Responses web_search tool to Antigravity web_search requestType envelope and googleSearch
- Map Google groundingMetadata to Responses web_search_call output item and url_citation annotations
- Buffer streaming text deltas while awaiting groundingMetadata so web_search_call strictly precedes message in SSE events and response.completed.output
- Derive search stream mode from effective translated request (requestRawJSON) to accurately support model aliases and rewrites
- Calculate streaming and non-streaming URL citation Unicode character (rune) offsets on full accumulated text, eliminating multi-byte CJK truncation and clamping
- Prioritize models.json native_capabilities.web_search explicit false as absolute veto before checking dynamic Antigravity probe capability
- Isolate Antigravity web search gating to Antigravity-specific model capabilities
- Suppress native googleSearch in Antigravity chat fallback when tools are mixed with function declarations
- Support Responses allowed_tools tool_choice containing web search and concatenate multi-part text queries
- Resolve Vertex Search grounding redirect URLs to target destination URLs in Antigravity executor
- Add comprehensive unit test coverage for stream/non-stream translation, late grounding, CJK offsets, model aliases, mixed tools, allowed_tools, and URL resolution
- Add logic to cap long OpenAI Responses tool names to 64 characters while retaining the most identifying portion.
- Implement disambiguation for truncation-induced collisions by appending suffixes to ensure unique tool names.
- Prevent truncated names from overlapping with namespace-less local names or previously declared aliases.
- Refactor declaration processing logic to consistently apply capping and disambiguation across requests and replayed history.
- Extend test coverage to validate correct capping, disambiguation, and namespace recovery for tool calls in streaming and non-streaming scenarios.
Closes: #5856
- Refactor input token computation to account for inclusive and exclusive cache scenarios, supporting fields like `input_tokens`, `total_input_tokens`, and `prompt_tokens`.
- Add support for distinguishing and handling `cache_read_input_tokens` and `cache_creation_input_tokens` separately.
- Update Claude translator logic to ensure correct token usage attribution, including fallback mechanisms for missing fields.
- Extend test coverage to validate various cache hit, partial, and write scenarios across streaming and non-streaming modes.
Closes: #5854
- Refine finish reason assignment logic in OpenAI interactions to account for incomplete states (`length`, `content_filter`) and tool call indices.
- Implement changes to normalize tool call indexing to ensure contiguous 0-based indexing.
- Add `status` and `incomplete_details` fields to OpenAI response payloads for enriched status handling.
- Improve logic for handling `generation_config` fields during interactions request conversion.
- Add robust defaulting for missing usage metrics in response payloads.
- Ensure comprehensive test coverage for all enhanced scenarios.
Closes: #5851
- Remove unexpected id parameter from function_result in Claude and OpenAI Chat Interactions request translators to comply with Google Interactions API schema and fix HTTP 400 (Unknown parameter 'id' at 'input[N]').
- Remove redundant call_id parameter from function_call in Claude Interactions request translator to match schema requirements.
- Propagate is_error flag between Claude tool_result and Interactions function_result.
- Align Devin executor to prioritize call_id for function_result steps.
- Add regression tests covering parameter schemas and end-to-end executor request generation.
Closes: #5828
- Introduced `meta` as a supported OAuth channel for model aliasing and request-scoped error rules.
- Added alias config to rename `muse-spark-1.3` as `muse-latest` with enforced mapping and fork behavior.
- Extended error rules for `meta` to classify "context_length_exceeded" responses with status `400` as `stop` actions.
- Updated examples and synthesizer logic to merge exclusions and apply Meta-specific mechanisms.
- Introduced `MetaExecutor` to support native Meta (Muse Code) API operations, including `Execute`, `ExecuteStream`, and `CountTokens`.
- Added logic to handle Meta-specific `responses` endpoint, including request preparation, enriched authentication, header management, and response translation.
- Implemented token counting functionality with custom `CountTokens` logic.
- Updated `meta_executor` to process streamed and non-streamed responses, maintaining compatibility with the Codex schema.
- Added utility functions to handle events, errors, and enriched metadata common to Meta API requests.
- Extended `models.json` to include Meta's updated configuration with additional thinking levels like "minimal" and "max."
- Introduced comprehensive tests to validate Meta-specific execution, streaming, header assignments, and response formatting.
- Introduce `hasValidToolCallArguments` to validate tool call arguments in OpenAI response translation.
- Update `effectiveOpenAIFinishReason` logic to consider argument validity for determining finish reasons like `tool_calls` or `length`.
- Extend finish reason assignment during response parsing to ensure robust handling of invalid or truncated tool call arguments.
- Add comprehensive test coverage for various tool call argument scenarios and finish reason outputs.
Closes: #5836
- Introduce `promoteOpenAIResponsesReasoningTextToSummary` to move `reasoning_text` parts from content to summary when the summary is empty.
- Clear `reasoning.content` to comply with Codex schema constraints (`maxItems: 0`).
- Implement safeguards for handling cleartext reasoning, preserving valid encrypted content, and stripping invalid `encrypted_content`.
- Add comprehensive tests to validate behavior for promoting reasoning texts, preserving existing summaries, and ensuring sanitization rules.
Closes: #5825
- Introduce logic to convert orphaned or unpaired `function_call_output` into user text instead of mismatched tool responses.
- Add `buildOpenAIResponsesStandaloneToolOutputTextParts` and `appendStandaloneResponsesToolOutputAsUser` helpers for managing orphaned outputs.
- Update OpenAI, Gemini, and Antigravity translators to ensure unmatched tool outputs are surfaced correctly.
- Add comprehensive test cases for both Gemini and OpenAI translator workflows to validate behavior for orphan function outputs and explicit call IDs.
Closes: #5831
- Introduce `sanitizeGeminiInteractionsUnsupportedInputIDs` to remove `id` fields from Gemini interactions `input` and `content` items.
- Update execution logic to apply the sanitization process before upstreaming payloads.
- Add tests to ensure payloads are correctly sanitized and validate `call_id` pairing for Gemini interactions.
Closes: #5828
- Add tests to verify that runtime hooks, auth modifications, and related updates do not block on unrelated operations such as antigravity probes or plugin virtual models.
- Introduce detailed scenarios, e.g., stale disables, batch operations, and conflict handling during concurrent auth updates.
- Refactor locking mechanisms to avoid unnecessary blocking during hooks and model registrations.
Fixed: #5813Closes: #5773
- Add `ForcedProvider` and `AuthID` fields to `HostModelExecutionRequest` and `ModelExecutionRequest`.
- Pin credentials via `WithPinnedAuthID` and forward forced provider options during model execution and streaming.
- Forward forced provider and auth ID from plugin host callbacks to model execution handlers.
Closes: #5814
