full-pm-cycle
Pass
Audited by Gen Agent Trust Hub on Jul 31, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes a plan-and-execute architecture to call various internal sub-skills (e.g., /market-intelligence, /prd-gen) for different stages of the PM lifecycle.
- [PROMPT_INJECTION]: The workflow processes user-provided inputs and data from multiple sub-skills, creating a surface for indirect prompt injection (Category 8). 1. Ingestion points: User input and current-workflow.json state files. 2. Boundary markers: None explicit in the skill instructions. 3. Capability inventory: Invokes external sub-skills and writes files to the project directory. 4. Sanitization: No explicit sanitization of input data is defined. This surface is characteristic of multi-stage AI agent workflows.
- [SAFE]: No evidence of hardcoded credentials, malicious persistence, or data exfiltration was found. The skill operates strictly within the project's documentation directories.
Audit Metadata