grilling

Pass

Audited by Gen Agent Trust Hub on Sep 11, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill instructions specify that the agent should proactively search the environment (file system, tools) for facts using sub-agents. This creates a surface for indirect prompt injection if the environment contains malicious data.
  • Ingestion points: File system and tool environment.
  • Boundary markers: None specified.
  • Capability inventory: File system and tool access are explicitly mentioned.
  • Sanitization: None specified.
  • [NO_CODE]: The skill consists of markdown instructions and does not include any scripts or executable code.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 11, 2026, 02:04 AM
Security Audit — agent-trust-hub — grilling