leadership-workflow
Pass
Audited by Gen Agent Trust Hub on Aug 2, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides a comprehensive framework for project management and leadership roles. It uses instructional templates to guide the agent through software development lifecycles.
- [DATA_EXPOSURE]: No hardcoded credentials, API keys, or sensitive local file paths (such as SSH keys or AWS configurations) were detected in the instructions.
- [REMOTE_CODE_EXECUTION]: The skill does not contain any commands to download external scripts or install third-party packages. It relies entirely on standard built-in tools for task management and memory.
- [PROMPT_INJECTION]: The instructions focus on structured output and task delegation. There are no attempts to override system safety filters or bypass core agent constraints.
- [INDIRECT_PROMPT_INJECTION]: The skill processes project requirements and user feedback as input. While it lacks explicit sanitization boundary markers, the input is used for high-level planning and delegation to sub-agents, which is the primary intended function and poses no significant security threat.
Audit Metadata