multi-brain
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill does not contain any malicious code, external downloads from untrusted sources, or data exfiltration attempts. It implements a standard brainstorming protocol using natural language instructions to provide multi-perspective analysis.
- [INDIRECT_PROMPT_INJECTION]: The skill creates a reasoning chain based on user-provided requests, which constitutes an attack surface for indirect prompt injection. However, the risk is mitigated by the structured nature of the workflow and the primary use case of synthesizing viewpoints.
- Ingestion points: User requests provided to the brainstorm workflow (SKILL.md).
- Boundary markers: None explicitly defined to separate user input from the perspective-taking logic.
- Capability inventory: Text generation, markdown report creation, and code generation with file system write permissions as part of the agent's expected output (SKILL.md).
- Sanitization: No explicit validation or filtering of user requests is performed before processing.
Audit Metadata