The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes content from SKILL.md files located in the skills/ directory and interpolates this data into README.md and .claude-plugin/marketplace.json.
Ingestion points: Reads name and description frontmatter from all files matching skills/*/SKILL.md.
Boundary markers: The instructions do not define delimiters or specific boundary markers for the data being interpolated into the target files.
Capability inventory: The skill performs local file system read and write operations.
Sanitization: There are no instructions for sanitizing, escaping, or validating the strings retrieved from external skills before they are written to central configuration and documentation files.

antv-skills-maintainer