finishing-a-development-branch

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION] (LOW): The skill executes arbitrary project test suites such as npm test, cargo test, and pytest. This is a primary function of the skill, but it involves executing code present in the local workspace.
  • [COMMAND_EXECUTION] (LOW): The skill uses git and the GitHub CLI (gh) to push code and create Pull Requests, which involves network interaction and credential usage.
  • [COMMAND_EXECUTION] (LOW): The skill processes project metadata like commit messages to populate Pull Request templates, creating a surface for indirect prompt injection. Ingestion points: git commit history and command output. Boundary markers: Absent for interpolated text in PR bodies. Capability inventory: Local file system access and shell command execution. Sanitization: None identified beyond standard agent summarization.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:15 PM