documentation-patterns
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted ADVPL/TLPP source code to generate technical documentation, creating a potential surface for malicious instructions to influence the agent's logic.\n
- Ingestion points: Source code files targeted for documentation (SKILL.md, reference.md).\n
- Boundary markers: The instructions do not define specific delimiters to isolate the code being analyzed from the agent's internal reasoning.\n
- Capability inventory: The skill performs file reading, accesses git log and git blame for metadata, and uses tools like grep to map function dependencies (reference.md).\n
- Sanitization: There is no mention of sanitizing or filtering the content of the source code before it is used to populate documentation templates.
Audit Metadata