using-git-worktrees
Pass
Audited by Gen Agent Trust Hub on May 22, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill performs downloads of project dependencies from well-known public registries through standard tools like
npm,pip,poetry, andgo, which is expected behavior for its stated purpose.\n- [COMMAND_EXECUTION]: Automates project setup and baseline verification by executing shell commands such asnpm install,cargo build, and various test runners (e.g.,pytest,npm test) based on detected project configuration files.\n- [PROMPT_INJECTION]: The skill exposes an indirect prompt injection surface by ingesting data from project-local files that may be under the control of an external actor.\n - Ingestion points: The skill reads
CLAUDE.mdto identify user preferences for worktree directories.\n - Boundary markers: No boundary markers or special delimiters are used to isolate the data read from
CLAUDE.mdfrom the rest of the agent's instructions.\n - Capability inventory: The skill possesses the capability to execute shell commands, install software packages, and perform git repository modifications.\n
- Sanitization: No sanitization, escaping, or validation logic is applied to the data retrieved from the configuration file before use.
Audit Metadata