field-keywords
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFENO_CODEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted data in the form of 'research notes' for domain detection and chapter mapping.
- Ingestion points: The 'Usage' section describes an agent process that ingests external research notes and file content to calculate relevance scores.
- Boundary markers: The instructions do not define specific delimiters or security warnings to ignore instructions embedded within the research notes.
- Capability inventory: The skill contains no scripts or tool definitions; it only provides data and logic for the agent's internal classification, which minimizes the potential for abuse.
- Sanitization: No sanitization or escaping of external content is specified.
- [NO_CODE]: The skill consists entirely of markdown documentation and JSON data blocks. It does not include any executable Python, Node.js, or Shell scripts.
- [SAFE]: No security threats were identified. The skill does not attempt to access credentials, perform network requests, or execute commands.
Audit Metadata