feat(action): GitHub Marketplace Action — composite action.yml (#14)#22
Merged
Conversation
Root action.yml makes agenteval a first-class GitHub Action (matching promptfoo):
installs agenteval from PyPI and runs `agenteval ci` as a build gate, failing
when pass-rate/regression breaches thresholds.
- action.yml: composite action; inputs mirror the ci command (suite, agent,
min-pass-rate, max-regression, baseline, format, output, parallel, version,
python-version) + Marketplace branding. Inputs are passed via env (NOT
interpolated into the run body) to avoid GitHub Actions shell-injection; every
expansion is quoted.
- README: a GitHub Action usage section.
PyPI-based (no ghcr needed) → fully functional today. A prebuilt ghcr.io image
for faster/hermetic runs is an optional follow-up (owner-gated on org-namespace
publish, #96) — documented in OWNER-ACTIONS, not blocking this action.
Tests: test_action_yml.py (valid composite Marketplace action; required
suite/agent inputs + defaults; steps invoke `agenteval ci` via env, no `${{ }}`
injection in the script body). pytest 3/3, ruff clean.
Closes #14.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_019tXZpN29akdmG8AEjgSZwk
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Closes #14.
What: a root
action.ymlmakes AgentEval a first-class GitHub Action (matching promptfoo) — installs agenteval from PyPI and runsagenteval cias a build gate, failing when pass-rate/regression breaches thresholds.How:
action.yml— composite; inputs mirror thecicommand (suite, agent, min-pass-rate, max-regression, baseline, format, output, parallel, version, python-version) + Marketplace branding. Inputs passed viaenv:(not interpolated into therunbody) to avoid GitHub Actions shell-injection; every expansion quoted.PyPI-based → functional today, no ghcr needed. A prebuilt
ghcr.ioimage (faster/hermetic) is an optional follow-up, owner-gated on the org-namespace publish (#96) — documented in OWNER-ACTIONS, not blocking this action.Tests:
test_action_yml.py(valid composite Marketplace action; required inputs + defaults; steps invokeagenteval civia env with no${{ }}injection in the body). pytest 3/3, ruff clean.