feat: add tool-call scorer - #55
Merged
royalpinto007 merged 2 commits intoSep 17, 2026
Merged
Conversation
Parses output as JSON and requires a tool call: name on the allowedTools allowlist and arguments as a plain object, with optional per-tool argument schemas reusing validate(). Closes AgentPostmortem#39
|
Merged. Thanks a lot, @sharadvc. Evalgate gets better every time someone outside the repo digs in like this. If you find Evalgate useful, consider dropping a star. It helps more people find it. Happy to see more from you whenever you have the time. |
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
2 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Fixes #39
Agent evals need "did the model emit a callable action" —
json-schemahas no function-call semantics.Change
src/scorers/tool-call.ts(parse-then-validate shape likejson-schema.ts).tool-calltype inregistry.tsbuiltinScorers.{name, arguments}:namemust be in theallowedToolsallowlist, andargumentsmust be a plain object. Optional per-toolschemasreuse thevalidate()helper.[Unreleased].Acceptance
ctx('{"name": "get_weather", "arguments": {"city": "London"}}')withallowedTools: ["get_weather"]passes.{"name": "rm_rf", ...}or non-JSON output fails.Full suite: 113/113 pass (
npm test);tsc --noEmitandeslintclean.This contribution was made with AI assistance (Claude).