workflow
Pass
Audited by Gen Agent Trust Hub on Nov 19, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill operates as a read-only utility. It scans the local filesystem using
os.walkto identify project structures and configuration files but does not modify any files. - [SAFE]: No network activity was detected. The script does not import any networking libraries (e.g.,
requests,urllib) and relies entirely on local file parsing and environment checks viashutil.which. - [SAFE]: The tool follows a separation of concerns where it merely constructs command-line argument arrays (
Step.command). It prints these to standard output as JSON for the agent to use but does not invoke sub-processes or shell commands itself. - [SAFE]: Secure parsing practices are observed, such as using
xml.etree.ElementTreefor project files and restricted file discovery that skips sensitive directories like.gitand.venv.
Audit Metadata