research

Pass

Audited by Gen Agent Trust Hub on Sep 6, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process information from untrusted external primary sources (official docs, specs, source code, APIs) and project-specific files.
  • Ingestion points: The process involves reading external documentation and local files such as CONTEXT.md and LINKS.md (SKILL.md).
  • Boundary markers: There are no instructions or delimiters defined to prevent the agent from following malicious commands that might be embedded in the external research sources.
  • Capability inventory: The skill possesses file-write capabilities, specifically writing markdown research artifacts to the .agents/projects/ directory (SKILL.md).
  • Sanitization: No sanitization, validation, or filtering mechanisms are specified for the external content before it is cited or written to the project artifacts.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 6, 2026, 01:25 PM
Security Audit — agent-trust-hub — research