finishing-a-development-branch
Pass
Audited by Gen Agent Trust Hub on Apr 20, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes local development tools and commands, including language-specific test runners (
npm test,cargo test,pytest,go test), Git for branch management, and the GitHub CLI (gh) for PR creation. These operations are standard for the skill's intended purpose. - [DATA_EXFILTRATION]: The skill performs network operations via
git pushandgh pr create. These actions are scoped to the project's configured remote repository and are expected behaviors for a development-focused skill. - [PROMPT_INJECTION]: No prompt injection patterns, role-play instructions, or system prompt extraction attempts were detected. The instructions follow clear, logic-based steps for task completion.
- [COMMAND_EXECUTION]: Potential Indirect Prompt Injection Surface:
- Ingestion points: Reads the project's test suite and Git branch names (SKILL.md).
- Boundary markers: Absent.
- Capability inventory: Subprocess execution for tests, Git operations, and GitHub CLI calls (SKILL.md).
- Sanitization: Absent. The skill relies on the existing environment and project configuration for command execution safety.
Audit Metadata