plan-grilling

Pass

Audited by Gen Agent Trust Hub on Sep 22, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The instructions encourage the agent to look up facts by exploring the filesystem and tools rather than asking the user. This creates an attack surface where the agent could read a file or tool output containing hidden instructions designed to manipulate its behavior.
  • Ingestion points: Instructions in SKILL.md to look up facts via the filesystem and tools.
  • Boundary markers: None specified to delimit external data from instructions.
  • Capability inventory: The skill leverages the agent's existing filesystem and tool access.
  • Sanitization: No specific sanitization or validation of the retrieved facts is mentioned.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 22, 2026, 02:55 PM
Security Audit — agent-trust-hub — plan-grilling