The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local Node.js scripts (scaffold-state.mjs, export-documents.mjs) located in the {job_hunt_skills_root} directory to manage document state and export cover letters to formats like PDF and Word.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it processes untrusted job descriptions and company research. 1. Ingestion points: Job posting content and company research reports are read into the context during the drafting phase in SKILL.md. 2. Boundary markers: There are no explicit delimiters or 'ignore' instructions for the interpolated external data. 3. Capability inventory: The skill can write files to the local filesystem and execute Node.js scripts. 4. Sanitization: It uses a claim-check process to verify that drafted content is supported by the user's source documents.

cover-letter