worktree
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill relies on a project-specific CLI tool named
noodleto manage git worktrees and execute shell commands within them vianoodle worktree exec. - [INDIRECT_PROMPT_INJECTION]: The skill handles branch names and shell commands provided by the user or other agents, creating a potential vector for malicious input. Ingestion points: The
<name>and<command...>arguments inSKILL.mdare passed to shell commands. Boundary markers: No delimiters or boundary markers are defined to isolate untrusted input from the execution environment. Capability inventory: The skill has the capability to execute arbitrary shell commands and perform git operations. Sanitization: No sanitization or validation of the input arguments is mentioned in the documentation.
Audit Metadata