nonlooped/ContextLevy
Estimate AI agent context overhead introduced by a pull request diff.
View on GitHubTrust Signals
- Scorecard Score
- not yet scored
- Maintenance Recency
- Stale
- License
- None
Inputs
| name | description | required | default |
|---|---|---|---|
| github-token | Fallback token with pull-requests: write when GitHub App auth is not configured | no | — |
| create-check | Publish a GitHub Check Run named ContextLevy (requires checks: write) | no | true |
| upload-sarif | Upload SARIF for Code Scanning annotations (requires security-events: write) | no | true |
| app-client-id | ContextLevy GitHub App ID. Falls back to CONTEXTLEVY_APP_ID or CONTEXTLEVY_APP_CLIENT_ID env var. | no | — |
| app-private-key | ContextLevy GitHub App private key PEM. Falls back to CONTEXTLEVY_APP_PRIVATE_KEY env var. | no | — |
| app-installation-id | Optional installation ID override. Falls back to CONTEXTLEVY_APP_INSTALLATION_ID env var. | no | — |
Outputs
| name | description |
|---|---|
| total-estimated-tokens | Total estimated AI-context tokens added by this PR (integer as string) |
| analyzed-file-count | Number of changed files included in the estimate (integer as string) |
| token-source | Auth source used for the run: app, github-token, or GITHUB_TOKEN |
| estimation-mode | Token estimation mode used for this run: simple or tokenizer |
| risk-level | Aggregated PR context risk level: Low, Medium, High, or Critical |
| check-conclusion | GitHub Check Run conclusion when create-check is enabled |
| badge-url | shields.io badge URL for PR context risk |
| badge-markdown | Markdown image tag for PR/README badges |
| sarif-path | Path to the generated SARIF file in the workspace |
| sarif-uploaded | Whether SARIF was uploaded to GitHub Code Scanning (true/false) |