git-worktree-manager

Pass

Audited by Gen Agent Trust Hub on Sep 8, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides standard Git and shell command instructions (e.g., git worktree, cp, npm install, createdb) intended for local project management and parallel development workflows.
  • [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to read and process local project files such as .env and package manager lock files (pnpm-lock.yaml, package-lock.json). While these files represent an external data ingestion surface, their use is limited to standard development setup tasks.
  • Ingestion points: Environment files (.env, .env.local) and lock files.
  • Boundary markers: None specified.
  • Capability inventory: File system read/write (copying env files), shell command execution for Git and package managers.
  • Sanitization: None specified.
  • [DATA_EXFILTRATION]: The skill describes copying sensitive environment files (.env, .env.local) between local directories. This operation is performed within the local file system to maintain environment parity between worktrees and does not involve network exfiltration.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 8, 2026, 06:55 AM
Security Audit — agent-trust-hub — git-worktree-manager