The Agent Skills Directory

[COMMAND_EXECUTION]: The skill invokes local Python scripts (scripts/collect-site-metrics.py and scripts/validate-readme.py) to perform structural analysis of the repository and ensure the generated README adheres to best practices. These scripts are limited to the project scope and do not perform network operations.
[PROMPT_INJECTION]: The skill facilitates an indirect prompt injection surface as it summarizes repository content to generate project overviews. 1. Ingestion points: SKILL.md Step 1 and Step 5 (reading mkdocs.yml and documentation files in /docs). 2. Boundary markers: Absent (no delimiters used when processing external project content). 3. Capability inventory: Local command execution and file writing. 4. Sanitization: Absent (the agent directly summarizes the ingested data into the final README output).

readme-generator