The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses a local Python script scripts/publish.py to perform file system operations and template rendering. It also directs the agent to execute standard development commands using git and the GitHub CLI (gh) to initialize repositories and push code.
[INDIRECT_PROMPT_INJECTION]: The skill processes the content of external SKILL.md files to extract metadata for repository generation, which constitutes an ingestion surface for untrusted data. • Ingestion points: The scripts/publish.py script reads the SKILL.md file from a user-provided directory path. • Boundary markers: No specific boundary markers or 'ignore' instructions are used when interpolating extracted metadata into generated templates. • Capability inventory: The script has the capability to read files, write files, create directories, and recursively delete directories (shutil.rmtree). • Sanitization: Extracted strings (name, description) are used directly in template replacement without validation or escaping.
[EXTERNAL_DOWNLOADS]: The skill suggests an optional installation of a companion validation tool from the author's own repository (psylch/better-skills) using npx.

better-skill-publish