stakeholder-management
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- [SAFE] (SAFE): Analysis of the provided files revealed no indicators of prompt injection, data exfiltration, or unauthorized command execution. The skill follows standard practices for role-play and domain-specific grounding.\n- [Indirect Prompt Injection] (SAFE): The skill is configured to ingest and prioritize content from reference files. While this creates a surface for data ingestion, the skill currently lacks sensitive capabilities (like network or file-system write access) that would make such a surface exploitable. Evidence:\n
- Ingestion points: references/patterns.md, references/sharp_edges.md, references/validations.md\n
- Boundary markers: Absent (the skill instructions direct the agent to consult files without specifying delimiters)\n
- Capability inventory: No subprocess calls, network operations, or file-system modifications detected in the provided scripts.\n
- Sanitization: Absent
Audit Metadata