finishing-a-development-branch
Pass
Audited by Gen Agent Trust Hub on Sep 26, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests and displays output from external test runners (e.g., npm test, pytest) during the verification step, which is a potential surface for indirect prompt injection if the project codebase contains malicious test results. * Ingestion points: Test command output from Step 1 in SKILL.md. * Boundary markers: No specific delimiters or instructions are used to separate test output from agent instructions. * Capability inventory: The skill utilizes git commands for merging, pushing, and cleaning up worktrees, as well as the GitHub CLI for creating pull requests. * Sanitization: There is no evidence of filtering or sanitizing the test runner output before it enters the agent context.
Audit Metadata