create-branch
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes local system commands using
git,gh(GitHub CLI), andglab(GitLab CLI) to manage branches and link issues. These operations are within the expected scope of a branch management tool. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user input to generate branch names and issue references.
- Ingestion points: User-provided branch descriptions and issue numbers in
SKILL.md. - Boundary markers: Defined via naming conventions and logic in
rules/prefix-detection.mdandrules/branch-naming.md. - Capability inventory: File system access via
git, and network communication viaghandglabCLI tools. - Sanitization: The skill contains robust validation rules in
rules/branch-naming.md, including character blacklisting (rejecting~,^,:,?,*,[,],\,@{,..) and forced conversion to kebab-case, which mitigates the risk of shell command injection via the branch name.
Audit Metadata