technical-writing
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill workflow requires the agent to ingest and analyze potentially untrusted external data, including source code, API schemas, design documents, and commit histories. Malicious instructions embedded within these source materials (e.g., in code comments or document metadata) could potentially influence the agent's behavior during the documentation process.\n
- Ingestion points: The 'Gather Source Material' section specifies the collection of source code, existing documentation, design documents, user stories, API schemas (OpenAPI/Swagger), and commit histories from the environment or user.\n
- Boundary markers: The skill lacks instructions for the agent to use delimiters or specific boundary markers to isolate external data from its primary instructions, increasing the risk of instruction leakage.\n
- Capability inventory: The skill is designed for text analysis and documentation generation; it does not define any executable scripts or high-risk capabilities such as network operations or file system writes within the SKILL.md instructions.\n
- Sanitization: No sanitization or validation steps are included to filter or escape the content of the ingested source materials before processing.
Audit Metadata