The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs users to run wt config shell install followed by source ~/.zshrc. This modification to shell profile files constitutes a persistence mechanism that can alter the user's environment or execute code automatically in every new shell session.
[EXTERNAL_DOWNLOADS]: Recommends the installation of max-sixty/worktrunk/wt via Homebrew. This repository is not associated with a trusted organization or well-known service, introducing a dependency on unverified external code that could pose a supply chain risk.
[PROMPT_INJECTION]: The ralph worktree-fix command ingests PR review comments as context for applying fixes, creating a surface for indirect prompt injection.
Ingestion points: External GitHub PR review comments are read into the agent context via the ralph worktree-fix workflow.
Boundary markers: Absent; the instructions do not implement delimiters or warnings to ignore instructions embedded in the external comments.
Capability inventory: The skill can execute shell commands (git, gh, wt), modify local files, and perform network operations via the GitHub CLI.
Sanitization: No sanitization or validation of the untrusted comment content is described or implemented.

worktree-pr