finishing-a-development-branch
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill automates standard development workflows using shell commands for Git operations (e.g.,
git merge,git push,git worktree) and various language-specific test runners (e.g.,npm test,pytest,cargo test). These executions are scoped to the intended functionality of branch management. - [INDIRECT_PROMPT_INJECTION]: The skill processes and displays output from external commands such as test suite results and git status porcelain output. This establishes an attack surface where a malicious project could attempt to influence the agent's behavior through crafted command output.
- Ingestion points: Capture of test suite output (Step 1) and Git status output (Step 6) into the agent's context.
- Boundary markers: Absent; tool output is presented to the user/agent without specific delimiters or isolation instructions.
- Capability inventory: Local shell execution (git, npm, cargo, pytest, go), file system access via git commands, and network operations via
git pushto the configured origin. - Sanitization: No explicit sanitization or filtering of external command output is documented.
Audit Metadata