devils-advocate
Pass
Audited by Gen Agent Trust Hub on Nov 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest and analyze external documentation and artifacts, creating a potential surface for indirect prompt injection.\n
- Ingestion points: The "Interactive Grilling Mode" section in
SKILL.mdspecifies that the agent should "Inspect supplied artifacts and available sources" and "Run a blindspot pass over relevant source, documentation, tests, documented limits, and known failure modes."\n - Boundary markers: The instructions do not define specific delimiters or boundary markers to separate untrusted external content from the agent's core instructions.\n
- Capability inventory: The skill leverages the agent's ability to read and analyze files within the environment. It does not explicitly define or use tools for network communication or file modification within the skill body.\n
- Sanitization: There are no instructions for sanitizing or validating the content of the external artifacts before processing.\n- [NO_CODE]: The skill consists entirely of markdown instructions and YAML metadata without any accompanying scripts, executables, or package dependencies.
Audit Metadata