fix(collect-worker-interface): collect function schemas from engine::functions::info - #285
Merged
Merged
Conversation
…functions::info
engine::functions::list returns only a FunctionSummary
(function_id/worker_name/description) — the typed request/response schemas
live solely on engine::functions::info (FunctionDetail), keyed
request_schema/response_schema. The collector read schemas from ::list under
the never-present request_format key, so every function normalized to an empty
{} schema and the --assert-typed-schemas publish check failed for every worker
regardless of how its handlers were typed.
Enrich each target function's row from engine::functions::info before
normalizing, and read request_schema/response_schema (with a request_format
fallback). Adds unit coverage for the enrichment merge and the normalize path.
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
|
No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info⚙️ Run configurationConfiguration used: Organization UI Review profile: CHILL Plan: Pro Run ID: 📒 Files selected for processing (4)
📝 WalkthroughWalkthroughAdds ChangesTyped Schema Enrichment Pipeline
Estimated code review effort🎯 2 (Simple) | ⏱️ ~10 minutes Possibly related PRs
Suggested reviewers
Poem
🚥 Pre-merge checks | ✅ 4 | ❌ 1❌ Failed checks (1 warning)
✅ Passed checks (4 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches📝 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 |
Contributor
skill-check — worker0 verified, 22 skipped (no docs/).
Four for four. Nicely done. |
ytallo
added a commit
that referenced
this pull request
Jun 18, 2026
…290) Every published worker trigger showed empty invocation_schema/return_schema (rendered "unknown" in the registry). The publish pipeline collected trigger types from engine::triggers::list, which returns only a TriggerTypeSummary (id/worker_name/description) with no schemas, and build_publish_payload read the obsolete trigger_request_format/call_request_format keys. The typed schemas live solely on engine::triggers::info (TriggerTypeDetail), under configuration_schema (binding config) and request_schema (delivered payload). This is the trigger-side twin of the function-schema fix (#285), which enriched functions from engine::functions::info but left triggers behind. - collect_worker_interface.py: enrich each publishable trigger type from engine::triggers::info before normalizing (mirrors enrich_functions_with_schemas); warn (non-fatal) on triggers still publishing without a typed invocation_schema. - build_publish_payload.py: read configuration_schema -> invocation_schema and request_schema -> return_schema, with the legacy keys as fallback. - tests: trigger enrichment + normalize field-mapping coverage. Trigger schemas are a warning, not a hard gate: some trigger types take no binding config (e.g. iii-directory's directory::*::on-change) and are legitimately schema-less. Claude-Session: https://claude.ai/code/session_017CciKnWZzsLprzZ6XFHZJ2
This was referenced Jul 3, 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
1 participant
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.
Problem
The publish-time
--assert-typed-schemascheck (collect_worker_interface.py) fails for every worker, reporting allrequest_schema/response_schemaas untyped (empty{}). It surfaced when releasingllm-router/v0.2.3, whose handlers are in fact correctly typed.Root cause
The collector gathers each function's schema from
engine::functions::list, but that endpoint returns only aFunctionSummary—function_id,worker_name,description, with no schema fields. The typed request/response schemas are exposed solely byengine::functions::info(FunctionDetail), under the keysrequest_schema/response_schema.normalize_worker_interfacealso read the never-presentrequest_formatkey.Result:
details.get("request_format")→None→{}for every function of every worker, so the assertion can never pass against a live engine. The schemas are correctly extracted by the SDK and stored by the engine — only the collection queried the wrong endpoint/keys.Confirmed on a live engine:
functions::listreturns 3 keys (no schemas);functions::inforeturns the full typed schemas.Fix
collect_worker_interface.py: enrich each target function's row fromengine::functions::info(best-effort, per function) before normalizing.build_publish_payload.py: readrequest_schema/response_schema(info-API key names), falling back torequest_formatfor back-compat.Verification
--assert-typed-schemasexits 0 (was 1).Note
The reusable publish workflow runs at the release tag's ref, so this needs a new tag (e.g.
llm-router/v0.2.4) to take effect in the pipeline — re-running the old job re-runs the pre-fix script.Summary by CodeRabbit
New Features
Tests