dojo-andrew-carnegie

Pass

Audited by Gen Agent Trust Hub on Sep 22, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user questions to determine routing and generate expert responses.
  • Ingestion points: User queries provided at runtime (SKILL.md).
  • Boundary markers: The instructions do not define clear delimiters or "ignore embedded instructions" warnings for user-provided data.
  • Capability inventory: The skill only performs text generation and reads local files from its own distribution; it lacks network, file-write, or shell execution capabilities.
  • Sanitization: No input validation or sanitization of user content is performed before interpolation into the persona's reasoning flow.
  • [SAFE]: The skill uses local markdown files to define behavior and knowledge. No remote dependencies, obfuscated code, or unauthorized network operations were found during analysis.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 22, 2026, 03:31 PM
Security Audit — agent-trust-hub — dojo-andrew-carnegie