devils-advocate
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data which could contain malicious instructions designed to subvert the analysis process or influence agent behavior.
- Ingestion points: The skill accepts a list of evidence materials as input, which can include local files, directories, and URLs, as described in the 'Entrance Judgement' and 'Step 0' sections of
SKILL.md. - Boundary markers: The instructions do not define specific delimiters or instructions to the model to ignore potential commands or formatting within the ingested evidence materials.
- Capability inventory: The skill utilizes file-reading capabilities to ingest evidence and file-writing capabilities to generate reports (
da_findings.jsonandda_report.md). - Sanitization: There is no evidence of sanitization, filtering, or validation of the content retrieved from external sources before it is interpolated into the agent's context.
Audit Metadata