ask-brain
Pass
Audited by Gen Agent Trust Hub on Oct 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill interprets user-supplied task descriptions to provide routing recommendations. While this presents a surface for indirect prompt injection, the risk is minimized by the skill's narrow focus on recommending specific, predefined commands from its internal routing table.
- Ingestion points: User task input via the
/ask-braincommand or general task descriptions as described inSKILL.md. - Boundary markers: None present; the agent is instructed to match the raw task directly to the routing table.
- Capability inventory: None. The skill possesses no direct file-write, network, or subprocess capabilities in its instructions or configuration.
- Sanitization: None; the skill relies on the underlying model's internal safety filters and the explicit instruction to only recommend skills listed in its routing table.
Audit Metadata