diplomatic-negotiator
Pass
Audited by Gen Agent Trust Hub on Oct 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill instructions emphasize evidence-based analysis, professional boundaries, and safety protocols. It explicitly prohibits the fabrication of facts and the impersonation of individuals or institutions.
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied information and external data sources for analysis, which represents a potential attack surface for indirect prompt injection.
- Ingestion points: External data and user inputs described in the 'Required Input' section of
SKILL.md. - Boundary markers: The 'Security and Ethics' section contains explicit directives to not treat third-party content as trusted instructions and to guard against prompt injection attacks.
- Capability inventory: The skill is limited to read-only operations using tools like
Read,Grep,Glob, andWebSearch, with no write or execution permissions. - Sanitization: Guidelines require the agent to cross-verify all information and clearly label the confidence level of conclusions (e.g., 'verified' vs 'pending confirmation').
Audit Metadata