docs: fix staleness across agenteval docs#24
Merged
Conversation
- action.yml: install the correct PyPI package (agentevalkit, not agenteval) - docker/docker-compose.yml: worker uses --broker (not --redis-url); drop the unread AGENTEVAL_REDIS_URL env from both services - github-actions/with-comparison.yml: run writes JSON via --report (no --format/-o); compare takes positional stored run IDs, not --baseline/--current JSON files - github-actions/with-gates.yml: same run fix; --gate takes a single gate-policy YAML file with GatePolicy keys, current run second (gate applies to last target); removed gate-results.json parsing (compare signals pass/fail via exit code) - README.md: 10 -> 11 graders in all three spots; add the trajectory grader row Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01CKkfVA3iY5G3Q43Mg4Xsu2
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Fixes verified doc-staleness issues from the audit pass. Docs only — no source logic touched.
pip install agentevalkit[==$VERSION](the package name inpyproject.tomlisagentevalkit, notagenteval).--broker(verifiedcommands/worker.py), not--redis-url. Removed theAGENTEVAL_REDIS_URLenv from both services since nothing reads it.agenteval runhas no--format/-o; JSON output is--report PATH.agenteval comparetakes positional run IDs stored in the SQLite DB, not--baseline/--currentJSON files. Rewrote accordingly.--reportfix.compare --gatetakes a single gate-policy YAML file (GatePolicy keys:min_pass_rate,score_threshold,max_regressions,max_cost_increase_pct,max_latency_increase_pct) with the current run as the second id (gate applies to the last target). Removed thegate-results.jsonparsing steps — compare writes no JSON and signals pass/fail via exit code.trajectorywas added); updated all three "10 graders" counts to 11 and added a table row for thetrajectorygrader.🤖 Generated with Claude Code
https://claude.ai/code/session_01CKkfVA3iY5G3Q43Mg4Xsu2