multi-agent-workforce-planner
Pass
Audited by Gen Agent Trust Hub on Apr 19, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides structured guidelines and templates for decomposing large features into parallel agent workstreams. No malicious patterns, such as prompt injection or unauthorized data access, were detected in any of the provided files.
- [SAFE]: Agent roles (Explore, Plan, Bash, Edit) are clearly defined with appropriate scopes. The Bash agent's ability to run commands is documented with standard development examples like unit testing and version control operations, which are benign in this context.
- [SAFE]: The Python code snippets provided in the dependency analysis reference are utility functions for static code analysis (extracting imports and detecting cycles) and do not perform any sensitive operations or network requests.
- [SAFE]: The skill processes external feature specifications which could potentially contain indirect prompt injections. However, the skill instructions are purely organizational and do not provide mechanisms to execute arbitrary code or bypass safety protocols based on that data.
Audit Metadata