The Agent Skills Directory

[SAFE]: The skill provides legitimate observability functionality for tracking AI code review performance through official tools.
[COMMAND_EXECUTION]: Includes Bash scripts that execute the GitHub CLI (gh) to perform API queries for repository metrics. The use of variables and jq parsing aligns with standard developer workflows.
[EXTERNAL_DOWNLOADS]: Configures GitHub Action workflows using the actions/github-script@v7 action, which is an official and well-known resource from the GitHub organization.
[INDIRECT_PROMPT_INJECTION]: The skill interacts with external repository data during the metrics collection process.
Ingestion points: Pull request metadata, review states, and comment counts fetched from the GitHub API in SKILL.md.
Boundary markers: Not explicitly present in the data processing scripts, as the focus is on quantitative aggregation.
Capability inventory: Uses gh api for read-only data retrieval and core.summary for reporting in GitHub Actions.
Sanitization: Data is parsed as JSON and filtered using jq, which prevents raw external strings from being interpreted as instructions by the scripts.

coderabbit-observability