finishing-a-development-branch
Pass
Audited by Gen Agent Trust Hub on Oct 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to read
AGENTS.mdand other maintainer documentation to determine branch protection policies and workflow rules. This creates a surface where malicious instructions in these external files could influence agent behavior. - Ingestion points:
AGENTS.mdand repository maintainer documentation (Step 2). - Boundary markers: Absent. The instructions do not define delimiters or warn the agent to ignore embedded instructions in these files.
- Capability inventory: The skill utilizes shell commands for
gitoperations (checkout, pull, merge, push, branch deletion), GitHub CLI (gh pr create), and various test suites (npm,cargo,pytest,go test). - Sanitization: Absent. Information extracted from these files is used to decide between local merges or pull request workflows.
Audit Metadata