multi-brain-score
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill provides a structured methodology for decision-making and does not contain any executable code, network requests, or file system interactions.
- [INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface as it ingests user input for multi-perspective evaluation. Ingestion points: User requests are processed by the 'Brainstorm' perspectives described in SKILL.md. Boundary markers: Absent. Capability inventory: None (the skill defines a text-based reasoning protocol and lacks subprocess calls, network operations, or file-writing capabilities). Sanitization: Absent. As the skill lacks any dangerous capabilities, this surface presents no practical risk.
Audit Metadata