skills/ntan-wan/skills/roast/Gen Agent Trust Hub

roast

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided plans and decisions to drive a fact-finding process. It directs the agent to autonomously explore the environment, creating a surface where malicious instructions within the user input could potentially influence tool use.
  • Ingestion points: User input defining the plan, decision, or idea to be interrogated in SKILL.md.
  • Boundary markers: The instructions lack explicit delimitation or warnings to ignore commands that might be embedded within the user's input.
  • Capability inventory: The skill explicitly grants the agent the ability to 'dispatch a sub-agent' to query the 'filesystem, tools, etc.'
  • Sanitization: There is no mention of validation, filtering, or sanitization of the user-provided context before it is used to guide environmental exploration.
  • [EXTERNAL_DOWNLOADS]: The skill documentation mentions an external repository at github.com/mattpocock/skills as the source of its adaptation. This is an informative reference for attribution and does not involve automated downloading or execution of remote code.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 06:10 AM
Security Audit — agent-trust-hub — roast