task-router
Pass
Audited by Gen Agent Trust Hub on Oct 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied project context to determine task routing, which introduces a potential surface for indirect prompt injection.
- Ingestion points: As defined in
SKILL.md, the model interprets external "user/project context" and "purpose pointers". - Boundary markers: The framework mandates recording facts, unknowns, and explicit risk levels, which helps maintain context integrity during transitions.
- Capability inventory: The skill can trigger other specialized procedures such as
security-and-hardening,skillfinder-external-scout, andadversarial-idea-review. - Sanitization: Mitigation is built-in via a mandatory
REVIEWmode for validating changes and the required use of security-specific gates (app-security-gate,security-and-hardening) when crossing trust boundaries or handling sensitive data.
Audit Metadata