The Agent Skills Directory

[SAFE]: No security issues were detected. The skill performs local analysis of project files to extract metadata such as script counts and feature descriptions for documentation generation. All scripts use standard Python libraries, and no network exfiltration or dangerous command execution patterns were identified.
[INDIRECT_PROMPT_INJECTION]: The skill processes project-specific files (e.g., SKILL.md, README.md) to populate documentation templates. While this introduces a potential ingestion point for external data, the extracted content is limited to short descriptive fields and used solely for generating static documentation files, posing no significant risk to the agent's execution environment.
Ingestion points: The scripts/analyze_project.py script reads local .claude/skills directory contents and parses SKILL.md files within the target project.
Boundary markers: None explicitly used during the markdown template interpolation process.
Capability inventory: The skill has the capability to read local project files and write SVG files to disk if an output path is provided.
Sanitization: Purpose strings extracted from project files are truncated to a maximum of 200 characters.

landing-page