agent-hiring-panel
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides a procedural methodology for auditing and hiring AI agents using structured role specifications, interview rubrics, and probation plans. No malicious instructions, obfuscation, or unauthorized resource access patterns were detected.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external data such as vendor claims and backlog tasks. 1. Ingestion points: Required inputs include backlog tasks and vendor evidence gathered during research. 2. Boundary markers: The skill explicitly instructs users to define boundaries and 'never-do' constraints. 3. Capability inventory: This specific skill file does not define or use any active tool capabilities (file system, network, or code execution). 4. Sanitization: While no explicit sanitization is mentioned, the lack of active capabilities makes the processing of external data safe within the context of this skill.
Audit Metadata