use-worktree
Pass
Audited by Gen Agent Trust Hub on Aug 16, 2026
Risk Level: SAFECOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructions direct the agent to 'Run the repository's recorded setup', which involves executing arbitrary shell commands or scripts (such as setup.sh, Makefile, or language-specific installers) found within the repository context. \n- [REMOTE_CODE_EXECUTION]: Executing setup scripts from a repository allows for the execution of arbitrary code. If a repository is malicious or has been compromised, using this skill to 'setup' the worktree will trigger the execution of the code provided by that repository. \n- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection through malicious repository content. \n
- Ingestion points: Repository setup files within the target directory. \n
- Boundary markers: None. The instructions do not specify any validation, human review, or isolation for the setup scripts. \n
- Capability inventory: Full shell access to execute git commands and repository-specific setup scripts. \n
- Sanitization: None. The agent is instructed to run the setup scripts without verifying their contents or origins.
Audit Metadata