git-branch-cleanup
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses Git and GitHub CLI commands to automate repository maintenance, including branch and worktree deletion. It executes shell pipelines to parse command output and manage the lifecycle of local development environments.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted metadata from branch names, commit messages, and GitHub PRs which could theoretically contain malicious instructions.
- Ingestion points: The skill reads branch names and commit subjects via
git for-each-refand retrieves pull request metadata viagh pr listinSKILL.md(Steps 1 and 2). - Boundary markers: No explicit delimiters are used to wrap ingested text, though data is organized into tables for human review before action.
- Capability inventory: The skill possesses destructive capabilities including
git branch -Dandgit worktree removeas defined in theSKILL.mdworkflow (Step 6). - Sanitization: The risk is mitigated by a mandatory user confirmation step for both branch categories, requiring explicit approval before executing any destructive commands.
Audit Metadata