The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process OpenAPI specifications provided by users or external sources. While this presents a surface for indirect prompt injection where malicious instructions could be embedded in schema descriptions, the skill focuses on generating static documentation files, which minimizes the risk of autonomous agent subversion.
Ingestion points: Ingests OpenAPI specs via the Read tool as specified in the Instructions section of SKILL.md.
Boundary markers: Not explicitly present; the skill treats the specification content as data for documentation.
Capability inventory: Uses Bash(api:docs-*), Write, and Edit tools to generate the documentation website and guide files.
Sanitization: The skill instructions recommend auditing the specification for completeness and enriching it, but do not explicitly mention sanitizing strings to prevent prompt injection.
[COMMAND_EXECUTION]: The skill utilizes a restricted Bash tool with a prefix filter api:docs-*. This is a defensive configuration that limits the agent's ability to execute arbitrary shell commands to only those related to documentation generation.
[EXTERNAL_DOWNLOADS]: The skill references several well-known and trusted technology services for documentation generation and hosting, including Redocly, Stoplight, Swagger, GitHub Pages, Netlify, and Vercel. These references are appropriate for the skill's stated purpose and do not represent a security risk.

generating-api-docs