Agent Claude Code
Use only for an explicitly requested quick spike or throwaway prototype that proves a hypothesis without production hardening. Never use for durable code, shared state, infrastructure, or real bug fixes.
✨ AI-first dotfiles with chezmoi
This repository also configures its own agents. See what dotfiles tells them →
Agent Claude Code
Use only for an explicitly requested quick spike or throwaway prototype that proves a hypothesis without production hardening. Never use for durable code, shared state, infrastructure, or real bug fixes.
Agent Claude Code
Review an implementation plan from a fresh, skeptical context before execution. Check claims against the repository, identify missing dependencies and verification, and return prioritized corrections without changing files.
Agent Claude Code
Use when drafting, reviewing, or improving a prompt for an LLM or agent — especially when an unbiased, clean-context reviewer is needed. Operates in its own context window with the prompt-engineering skill preloaded, so it isn't anchored to the conversation that produced the prompt.
Agent Claude Code
Write clean, safe, readable shell scripts following best practices. Use proactively whenever tasks involve creating or modifying shell scripts, bash scripts, or installation scripts.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: