fix(server): honor explicit $skill invocations across providers - #8336
fix(server): honor explicit $skill invocations across providers#8336D3OXY wants to merge 3 commits into
Conversation
|
Important Review skippedAuto reviews are disabled on this repository. Please check the settings in the CodeRabbit UI or the ⚙️ Run configurationConfiguration used: Repository UI Review profile: CHILL Plan: Pro Plus Run ID: You can disable this status message by setting the Use the checkbox below for a quick retry:
Comment |
ApprovabilityVerdict: Not approved Macroscope's review found this PR not approvable — This PR introduces a cross-provider skill invocation pipeline that changes provider health checks and send-turn processing, including filesystem scans, CLI/API discovery, prompt transformation, and structured skill inputs. Its multi-provider scope and substantial new runtime logic exceed a simple self-contained fix and merit human review. Notes:
You can add or adjust custom eligibility rules. Learn more. |
4d30d52 to
24b1136
Compare
There was a problem hiding this comment.
Reviewed the new/changed Effect service code (provider/skillInvocations.ts, Layers/codexSkillInvocations.ts, Drivers/CursorSkills.ts, Drivers/GrokSkills.ts, adapters, contracts). Imports, schema-tagged errors, and dependency acquisition follow the conventions; one error-translation issue found in CodexAdapter.mapCodexRuntimeError.
Posted via Macroscope — Effect Service Conventions
a594a03 to
6426645
Compare
6426645 to
af57619
Compare
Problem
The composer flattened a selected skill chip into plain
$nametext. Provider adapters then guessed intent with a regex, so ordinary dollar-prefixed text could be treated as a skill and provider behavior differed.Fix
{ name, start, end }from web/mobile composers through durable drafts, queued sends, turn-start events, and provider requests. Timeline text stays canonical.Verification
git diff --checkpass.Addresses #7795.
Implemented with GPT-5.6 Codex in the T3 Code harness.