worktree
Pass
Audited by Gen Agent Trust Hub on Sep 8, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute standard command-line tools such as
gitandghto manage repository state. This includes creating and removing worktrees, fetching repository updates, and checking out PRs. - [INDIRECT_PROMPT_INJECTION]: The skill defines a workflow that processes externally controlled repository metadata, creating a potential surface for indirect prompt injection.
- Ingestion points: The agent reads and acts upon branch names, pull request identifiers, and commit messages (SKILL.md).
- Boundary markers: There are no specific delimiters or instruction-ignore markers defined for the data ingested from the git environment.
- Capability inventory: The skill possesses capabilities to execute shell commands for git operations and filesystem navigation.
- Sanitization: While the skill encourages meaningful naming, it does not mandate specific sanitization of input strings before they are passed as arguments to git commands.
Audit Metadata