domain-research

Pass

Audited by Gen Agent Trust Hub on Sep 23, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill's workflow involves browsing and extracting information from external, untrusted websites, which serves as an ingestion point for potentially malicious instructions embedded in web content.
  • Ingestion points: The skill instructs the agent to browse targeted URLs and verify sources directly in SKILL.md (Step 3) and references/authorities-and-frameworks.md (Direct Source Verification).
  • Boundary markers: There are no explicit instructions or delimiters provided to the agent to distinguish between the research task and instructions that might be present in the external data.
  • Capability inventory: The agent is granted the ability to browse the web and write research findings to markdown files on disk.
  • Sanitization: No sanitization or validation of the external content is performed before it is synthesized into the final document.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 23, 2026, 04:20 AM
Security Audit — agent-trust-hub — domain-research