finishing-a-branch

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes standard git CLI commands to perform repository management tasks, including git merge, git push, git branch -d, and git worktree remove. These commands are executed based on the user's selected workflow and the environment identified via git rev-parse.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external data from the local repository environment that could theoretically be manipulated to influence the agent's behavior:
  • Ingestion points: The skill reads output from git rev-parse, git status --porcelain, and git branch to identify the environment and list uncommitted files (SKILL.md).
  • Boundary markers: There are no explicit boundary markers or delimiters used when interpolating these values into terminal output or reports.
  • Capability inventory: The skill has the capability to perform destructive actions, such as git branch -D (forced delete) and git worktree remove --force (SKILL.md).
  • Sanitization: The skill does not demonstrate specific sanitization or escaping of branch names or file paths before they are displayed or processed in logic flows.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 01:19 PM
Security Audit — agent-trust-hub — finishing-a-branch