git-worktree-discipline
Pass
Audited by Gen Agent Trust Hub on Aug 14, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes several shell commands to interact with the Git repository and the local filesystem. This includes 'git worktree add', 'git rev-parse', and 'git check-ignore' found in 'reference.md'. These commands are used to manage workspace isolation and detect current repository states.- [EXTERNAL_DOWNLOADS]: When setting up a new worktree, the skill triggers automated dependency installation and testing. It uses standard package managers such as 'npm', 'pip', 'poetry', 'cargo', and 'go' to fetch code from official public registries based on project files like 'package.json' or 'requirements.txt'.- [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection because it performs actions based on untrusted data within the user's project.
- Ingestion points: The skill reads settings from 'AGENTS.md' or 'CLAUDE.md' (e.g., 'worktree-gate') and scans for project manifest files.
- Boundary markers: There are no explicit boundary markers or instructions to the agent to disregard instructions embedded within these external files.
- Capability inventory: The skill allows the agent to execute complex installation and test routines through subprocesses based on detected markers.
- Sanitization: The instructions do not define sanitization or validation steps for the contents of the configuration files before they influence the execution flow.
Audit Metadata