gt2gh
Pass
Audited by Gen Agent Trust Hub on Aug 16, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill defines a restricted operating environment for the gt2gh CLI tool, emphasizing preview-first actions and requiring the --apply flag for any mutations to GitHub or Git states. It explicitly mandates using internal/subprocess.Runner and fake executables for testing to prevent accidental production impact.
- [COMMAND_EXECUTION]: The skill instructs the agent to execute git, gh, and gt commands. These are scoped to the tool's core functionality for stack management. Safety measures include using atomic pushes and preventing automatic repairs of repository states.
- [EXTERNAL_DOWNLOADS]: The instructions suggest installing the tool via brew install shhac/tap/gt2gh. This targets a resource controlled by the skill's author.
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from local repository files and CLI tool outputs. The instructions mitigate risks by requiring strict non-interactive parsing, schema validation, and human-in-the-loop review for PR submissions. Ingestion points include repository docs and tool output; capabilities include git and gh CLI calls; sanitization is addressed via strict parsing requirements; explicit boundary markers are absent.
Audit Metadata