git-merge-expert-worktree

Pass

Audited by Gen Agent Trust Hub on Apr 1, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
  • [COMMAND_EXECUTION]: The skill facilitates the Git worktree lifecycle using shell commands such as git worktree, git merge, git push, and the GitHub CLI (gh).\n- [EXTERNAL_DOWNLOADS]: The skill utilizes pnpm install and git fetch to manage dependencies and synchronize repository state with remote sources.\n- [DATA_EXFILTRATION]: The instructions involve accessing sensitive local files like .env and .env.local to replicate the environment configuration into newly created worktrees.\n- [PROMPT_INJECTION]: The skill has an indirect prompt injection surface as it reads and processes external Git data, including branch names, commit messages, and file content markers.\n
  • Ingestion points: Metadata and contents from Git branches, pull requests, and merge conflict markers (SKILL.md).\n
  • Boundary markers: The instructions do not define specific boundary markers or safety instructions are provided to the agent to ignore potentially malicious embedded content within the Git data.\n
  • Capability inventory: The skill maintains capabilities for shell command execution and filesystem modification (git, pnpm, cp, ln) (SKILL.md).\n
  • Sanitization: There is no explicit requirement to validate or sanitize input received from the Git repository.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 1, 2026, 11:28 AM