simgit-worktrees
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides instructions for executing shell commands via 'simgit' and 'git' to handle workspace lifecycles, such as creating, removing, and pruning Git worktrees.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data including branch names and filesystem paths that are subsequently used in shell commands.
- Ingestion points: Branch names, commit hashes, and file paths are provided by the agent or user for worktree allocation in 'SKILL.md'.
- Boundary markers: The skill does not explicitly define markers to prevent the agent from misinterpreting these strings as executable instructions.
- Capability inventory: The skill utilizes filesystem and repository management capabilities provided by 'simgit' and 'git'.
- Sanitization: A mandatory identity check ('simgit doctor --json') is included to ensure the agent uses the intended binary.
- [DYNAMIC_EXECUTION]: The skill constructs shell commands at runtime by interpolating variables such as cleanup tokens and branch names.
- Evidence: Command assembly examples such as 'simgit remove --json --delete-branch "$branch"' are detailed in the instructions.
Audit Metadata