worktrees

Pass

Audited by Gen Agent Trust Hub on May 20, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to manage project files and branch states using git worktree, git merge, and package management tools (pnpm). It includes shell loops for monitoring status across multiple directory paths.
  • [EXTERNAL_DOWNLOADS]: The installation process for this skill involves a remote download of the skill definition file from a public GitHub repository.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it is designed to read and follow instructions from project-level coordination files (PLAN.md and STATUS.md).
  • Ingestion points: Reads PLAN.md and STATUS-{name}.md from the project root (SKILL.md).
  • Boundary markers: None identified; instructions in these files are treated as authoritative for task coordination.
  • Capability inventory: The agent has the ability to execute shell commands (git, pnpm), read files (cat), and write files (SKILL.md).
  • Sanitization: No explicit sanitization or validation of the content within the coordination files is performed before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
May 20, 2026, 07:45 PM
Security Audit — agent-trust-hub — worktrees