3-prd
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from
devpost/scope.mdanddevpost/learner-profile.mdto inform the product requirements document. This creates a surface for indirect prompt injection if those files contain malicious instructions. - Ingestion points:
SKILL.md(lines 43, 62). - Boundary markers: No specific delimiters or instructions to ignore embedded prompts in the source files are defined.
- Capability inventory: Writing and updating
devpost/prd.mdanddevpost/prd.html. - Sanitization: No explicit sanitization or validation of the ingested content is mentioned.
- [EXTERNAL_DOWNLOADS]: The skill generates an HTML document that may fetch the Mermaid library from a well-known CDN to render diagrams. This is a standard practice for creating interactive documentation.
Audit Metadata