agentkitai/AgentEval
Run an AgentEval suite as a CI gate — fail the build when the agent pass rate or regression breaches your thresholds.
View on GitHubTrust Signals
- Scorecard Score
- not yet scored
- Maintenance Recency
- Activelast commit Jul 8, 2026
- License
- MIT
Pinned Snippet
uses: agentkitai/agenteval@48d54c97915b20f2c182d12e47d88e6a0b09f3df # v0.7.0tags can be moved; commit SHAs can't. why a SHA?
Inputs
| name | description | required | default |
|---|---|---|---|
| suite | Path to the eval suite file (YAML/JSON). | yes | — |
| agent | Agent callable as 'module:func'. | yes | — |
| min-pass-rate | Minimum pass rate (0–1) to pass the gate. | no | 0.8 |
| max-regression | Maximum allowed regression percentage vs the baseline. | no | 10 |
| baseline | Baseline run ID for regression detection (optional). | no | "" |
| format | Output format: text | json | junit. | no | text |
| output | Write the report to this file (optional). | no | "" |
| parallel | Max concurrent cases. | no | 1 |
| version | agenteval version to install from PyPI (e.g. '1.2.3'); empty = latest. | no | "" |
| python-version | Python version to set up. | no | 3.12 |
Outputs
no outputs