feat: land six repaired cherry-picks (MiniMax, picker order, terminal guard, streamAborted, imageInput, droid docs) - #1744
Merged
Merged
Conversation
Port the codex-router #139 streamAborted metering marker: when an upstream stream dies after its 200 head was committed, the persisted usage attempt now carries streamAborted: true alongside the synthetic 502 terminal so the turn never meters as a success the client did not receive. - usage/log.ts: optional streamAborted on PersistedUsageAttempt, whitelisted in normalizeUsageAttempt (only the literal true marker survives the round trip; ordinary and legacy rows keep their exact shape). - relay.ts: set the marker on consumeForInspection's mid-stream read-failure path (failed/502) and on trackSseForRequestLog's read-failure path (incomplete/502). Client-cancel (499) semantics are untouched and never carry the marker. - tests: usage-log.test.ts gains persistence/backward-compat coverage; stream-aborted-marker.test.ts drives the relay paths end to end through addFinalRequestLog to the persisted JSONL row.
…er order Large routed catalogs (10-20+ models across several providers) have no supported way to control the Codex model-picker display order beyond the 5-slot subagentModels list: every non-featured routed row is emitted at the same flat priority, so the picker order is undefined and reshuffles on each catalog rebuild (ocx sync / service restart / upgrade). Add an optional, display-only config.modelPickerOrder: string[]. Listed routed <provider>/<model> slugs are shown in array order in the picker; unlisted rows and subagentModels-featured rows keep their positions. When unset, catalog priority is byte-identical to before (the codex-catalog golden oracle is unchanged). Display and spawn_agent candidacy are fully decoupled: modelPickerOrder rewrites only the Codex-visible `priority`, while each moved row records its natural priority in an OpenCodex-private catalog field (opencodex_spawn_priority) that effectiveSubagentRoster uses to pick candidates. The spawn_agent candidate set is therefore provably unchanged by any display reordering — even reversing every row. Codex ignores the unknown field (same as opencodex_catalog_kind), so this is purely a user-facing picker feature. Fixes #1649
Per-combo image/multimodal policy: default auto keeps the target intersection; disabled strips image from catalog modalities and rejects image-bearing requests (including previous_response_id expansions) with HTTP 400 before dispatch. Management API sparsifies only explicit disabled. GUI exposes a capability switch when every target supports images.
Fail closed for incomplete targets in comboImagesSupported; omit default imageInput auto from management GET/PUT responses; reject unresolved previous_response_id when images are disabled; cover disabled draft persistence and normalizedCombo fixture.
Restrict image-disabled detection to materialised Responses input after previous_response_id expansion so tool schemas and metadata cannot false- positive HTTP 400. Treat unresolved continuations by object identity, strip previous_response_id after a successful expand to avoid double replay on child dispatch, and align docs with capability-intersection wording.
…ai-chat providers The no-tool-call terminal continuation guard (#394) is bound to the anthropic adapter only. Self-hosted OpenAI-compatible gateways (GLM/Kimi-family, etc.) routed through openai-chat hit the same premature-completion pattern -- the model announces work but ends the turn without emitting a tool call -- yet never get the bounded re-ask, so they stop mid-work. Extend the guard to openai-chat, gated behind a new per-provider opt-in flag `terminalContinuationGuard`. Default behavior is unchanged: anthropic keeps the guard, and the many registry providers sharing the openai-chat adapter stay off unless a provider explicitly enables it (the suspicious-no-tool-stop heuristic in analyzeTerminalTurn was tuned on Anthropic turns, so opt-in is the conservative default). Fixes #1651
…e path Address CodeRabbit review on #1660: add server-level integration coverage that exercises the openai-chat guard flag through handleResponses/core.ts rather than only passing adapterName to guardTerminalEventStream. - unset/absent terminalContinuationGuard on an openai-chat provider => no continuation (exactly one upstream call). - terminalContinuationGuard: true => one bounded continuation (two upstream calls) and the recovered tool call is forwarded. The existing anthropic combo-attempt and routed-compaction exclusions are unchanged and still covered by the surrounding suite.
Contributor
|
✅ Deterministic PR hygiene checks passed. |
Contributor
|
Caution Review failedThe pull request is closed. ℹ️ Recent review info⚙️ Run configurationConfiguration used: Path: .coderabbit.yaml Review profile: ASSERTIVE Plan: Pro Plus Run ID: ⛔ Files ignored due to path filters (1)
📒 Files selected for processing (83)
📝 WalkthroughWalkthroughThis change adds MiniMax Code and MiniMax CLI support, combo image-input configuration, display-only model ordering, terminal continuation handling, stream-abort metering, self-launch argument construction, GUI controls, documentation, localization, and tests. ChangesRuntime and integration features
GUI and documentation
Estimated code review effort: 5 (Critical) | ~120 minutes Possibly related PRs
Suggested labels: Suggested reviewers: ✨ Finishing Touches 💡 1🛠️ Fix failing CI checks 💡
📝 Generate docstrings
🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
This was referenced Aug 15, 2026
Contributor
⏳ DRAFT
What to do
Automatic draft conversion failed (token cannot change draft status). Please convert this pull request to a draft manually. The required |
This was referenced Aug 15, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
6 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary\n\nLands six repaired cherry-picks from old draft PRs (triage: devlog/_plan/260815_old_draft_triage/). Contributor commits were cherry-picked with authorship preserved, then repaired per review findings; each repair branch was reviewed independently (PASS after one folded blocker):\n\n- #1664 MiniMax Code/CLI integration (+ shared standalone-aware self-launch argv helper, migrating 12 launcher call sites incl. update/job, update/notify, system-restart)\n- #1669 modelPickerOrder (+ fail-soft malformed-config normalization, docs fix)\n- #1660 opt-in terminal guard for openai-chat (+ explicit-false and combo/compaction exclusion tests, docs)\n- #1652 streamAborted usage marker (+ eager/WS path coverage, cancellation race fix)\n- #1165 combo imageInput control (+ tr/zh-TW locale keys, CLI round-trip, anti-double-expansion regression lock)\n- #1644 Factory Droid bridge guide (+ provider-id/contract repairs, EN+KO)\n\n## Verification\n\n- Remote full gates on ssh lidge @ b05c494 (this branch tip): 12259 pass / 11 skip / 0 fail (12270 tests, 781 files); typecheck, privacy:scan, lint:gui all green.\n- Independent repair review: GO-WITH-FIXES -> blocker (2 missed launcher sites) fixed in 2f276eb and re-verified PASS by the same reviewer.\n\n## Checklist\n\n- [x] Targets dev\n- [x] No token/secret logging added\n- [x] Regression tests included for every behavior change
Summary by CodeRabbit