code-planner
Pass
Audited by Gen Agent Trust Hub on Jul 17, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill uses
git status --shortto inspect the state of the local repository before planning or assigning work. This is a standard and safe operation for a development-oriented tool to ensure it is working with a clean or understood state. - [EXTERNAL_DOWNLOADS]: The documentation mentions using
npx skillsto add or update the skill from the author's GitHub repository (github.com/idaibin/aicraft). This is an administrative instruction for the user and does not involve the agent executing untrusted remote code during its normal operation. - [SAFE]: The skill implements strong boundary controls by requiring structured output contracts and task templates, which helps mitigate risks associated with processing untrusted codebase content.
Audit Metadata