using-git-worktrees
Pass
Audited by Gen Agent Trust Hub on Oct 4, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes various development-related shell commands including
git worktree,npm install,cargo build,pip install, andgo mod download. These operations are triggered automatically based on the detection of specific project files (e.g.,package.json,Cargo.toml). - [INDIRECT_PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it ingests untrusted data from the repository environment.
- Ingestion points: Reads instructions from
CLAUDE.mdand evaluates the local file system structure to determine workspace locations. - Boundary markers: The skill does not explicitly define delimiters for external data within its logic.
- Capability inventory: Capable of creating directories, modifying
.gitignore, and executing package manager installation scripts. - Sanitization: No specific sanitization logic is present for data read from
CLAUDE.mdor branch names used in path construction.
- Ingestion points: Reads instructions from
Audit Metadata