project-manager

Pass

Audited by Gen Agent Trust Hub on Jul 27, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes a local bash script (scripts/init-project.sh) to automate the creation of a project directory (_project/) and various markdown templates (e.g., charter.md, risks.md). The script is idempotent, meaning it does not overwrite existing user data, and is constrained to the directory specified by the user.
  • [SAFE]: No patterns of prompt injection, obfuscation, or credential harvesting were identified. The instructions promote disciplined project management practices, such as citing evidence for status updates and maintaining an append-only decision log.
  • [SAFE]: The skill maintains all project data locally within the user's workspace. There are no network calls or external data transfers observed in the provided scripts or instructions.
  • [SAFE]: References to companion skills (product-manager, strategy-board, critical-thinking) are limited to the author's own verified ecosystem, which is standard for specialized agent toolkits.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 27, 2026, 02:35 AM
Security Audit — agent-trust-hub — project-manager