mizcausevic-dev/OTel GenAI Fleet Summary
Walk OTel GenAI spans, surface governance findings, post summary on PR, fail on missing-model or unauthorized.
View on GitHubTrust Signals
- Scorecard Score
- not yet scored
- Maintenance Recency
- Stale
- License
- None
Inputs
| name | description | required | default |
|---|---|---|---|
| traces-dir | Directory containing *.json OTLP exports. | yes | — |
| allowed-models | Comma-separated whitelist of approved model identifiers. Spans using non-whitelisted models trigger 'unauthorized-model' (high). | no | — |
| error-rate-threshold | Fraction (0..1) of errored spans above which 'high-error-rate' fires per file. Default 0.05. | no | — |
| comment-on-pr | Post the Markdown summary as a PR comment. 'auto' (default) posts only on pull_request events. | no | auto |
| fail-on-high | Fail the run when any high-severity finding is present (default 'true'). | no | true |
| github-token | Token used to post the PR comment (defaults to GITHUB_TOKEN). | no | ${{ github.token }} |
Outputs
| name | description |
|---|---|
| total-files | Number of OTLP files analyzed. |
| total-spans | Total GenAI spans across the fleet. |
| high-findings | Count of high-severity findings. |
| unique-models | Number of distinct models in use. |