finishing-a-development-branch
Pass
Audited by Gen Agent Trust Hub on Feb 27, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill performs various shell-based operations using standard development tools, including Git for branch management and the GitHub CLI (gh) for Pull Request creation.
- [COMMAND_EXECUTION]: The skill triggers the project's own test suite (e.g., npm test, pytest, cargo test, go test), which involves executing code residing within the local repository.
- [PROMPT_INJECTION]: Indirect Prompt Injection Surface.
- Ingestion points: Local repository content (test files, source code) and git metadata (branch names, commit history).
- Boundary markers: None present.
- Capability inventory: The skill has the ability to execute shell commands, manage git branches, and interact with remote repositories via the GitHub CLI.
- Sanitization: The skill template for creating Pull Requests interpolates a variable directly into a shell command; agents using this skill must ensure proper escaping to prevent command injection if the title originates from untrusted data.
Audit Metadata