cover-letter
Pass
Audited by Gen Agent Trust Hub on Oct 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted external data, including job descriptions (often via URLs) and resumes, creating a surface for indirect prompt injection attacks.
- Ingestion points: External job description text, URLs, and resume content are gathered in Step 1 and processed throughout the workflow.
- Boundary markers: The instructions do not define clear delimiters or use "ignore embedded instructions" warnings when handling user-provided or web-fetched content.
- Capability inventory: The skill uses
cat > file << 'EOF'to write content to the local filesystem (Step 5). While this is a common operation for writing documents, it could be abused if the generated content contains malicious shell commands or control characters intended for the terminal. - Sanitization: There is no mention of sanitizing, escaping, or filtering the input data before it is mapped to requirements or used in the final document generation.
Audit Metadata