The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes standard git commands for worktree management and triggers local package managers (npm, pip, cargo, go) to set up the development environment. These operations are restricted to the local repository context.- [EXTERNAL_DOWNLOADS]: Performs network operations via git fetch and package managers to synchronize with remote repositories and install project dependencies from official registries.- [PROMPT_INJECTION]: The skill processes untrusted data from the Linear API (issue titles). It mitigates injection risks by implementing a slugification process in Step 5 that converts titles to lowercase and replaces special characters with hyphens before they are used in shell commands. Ingestion point: Issue titles from Linear:get_issue in SKILL.md. Boundary markers: Absent. Capability inventory: git, npm, pip, cargo, go. Sanitization: Step 5 slugification logic.

linear-worktree