healthcare
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted external data which constitutes a potential surface for indirect prompt injection.\n
- Ingestion points: Data enters the agent context through client briefs, social media page exports, and existing policies provided by the user as specified in
SKILL.md.\n - Boundary markers: The instructions lack explicit delimiters or 'ignore' instructions to isolate user-provided data from core logic.\n
- Capability inventory: The skill generates social plans and response protocols; however, its operations are restricted to read-only analysis and searching, limiting the impact of potential command injections.\n
- Sanitization: No input validation or sanitization routines are defined for the ingested client information.
Audit Metadata