durable-docs-update
Pass
Audited by Gen Agent Trust Hub on Sep 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted input from code comments, changed files, and specification documents to automatically update durable documentation. This creates a surface for indirect prompt injection where malicious content within those files could manipulate the agent's logic or the resulting documentation.\n
- Ingestion points: The skill resolves scope and reads content from code files, comments, and spec.md as described in Step 1 and Step 2.\n
- Boundary markers: The protocol does not specify the use of delimiters or 'ignore' instructions when reading file content, which may lead to the agent following instructions embedded in the data.\n
- Capability inventory: The skill allows for the modification, deletion, and movement of text within files and the invocation of multiple text-processing tools.\n
- Sanitization: There is no evidence of sanitization or escaping of the extracted comment text before it is processed or written back to the files.
Audit Metadata