site-brief
Pass
Audited by Gen Agent Trust Hub on Sep 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external websites and repository files to auto-draft business briefs, creating a surface for injection attacks.\n
- Ingestion points: Data is fetched from user-provided URLs (homepages, services, reviews) and repository contents (READMEs, CMS exports) using browsing tools or plain fetches.\n
- Boundary markers: The instructions do not mandate the use of delimiters or 'ignore' instructions for the untrusted content being ingested into the context.\n
- Capability inventory: The skill possesses file-writing capabilities (to
.agents/business-context.md) and the ability to perform network operations throughagent-browser.\n - Sanitization: The instructions lack specific requirements for sanitizing or filtering external content before it is interpolated into the drafting prompt.
Audit Metadata