knowledge-review

Pass

Audited by Gen Agent Trust Hub on Aug 1, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it is designed to ingest and obey instructions found in an external manifest file (KNOWLEDGE.md).
  • Ingestion points: The skill reads the KNOWLEDGE.md file at the start of each invocation to determine audit criteria.
  • Boundary markers: Absent; there are no instructions to the AI to delimit or ignore potentially malicious instructions within the manifest.
  • Capability inventory: The skill is primarily read-only, but its instructions give ingested content the power to override its logic, and it references other active skills for follow-up actions.
  • Sanitization: Absent; the manifest content is not validated or filtered before being adopted as binding logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 1, 2026, 02:57 PM
Security Audit — agent-trust-hub — knowledge-review