Automation

A desktop notification with a sound when Claude finishes a long answer or may need your OK

A pixel-art office in a local web page: every Claude session gets a desk, walks to the filing cabinet when it reads, types when it edits, flies paper planes when it messages, and shows a live Done-by clock

Before Claude builds anything, it re-reads your own CLAUDE.md rules and keeps a reference's structure without copying its look

Put text on your clipboard from the prompt, cleaned of terminal clutter.

Files you mark as yours: Claude cannot edit them until you hand them back.

Open a link or a file in its own window, straight from the prompt.

A chime when Claude finishes a long turn, so you can look away.

Hand Claude the picture on your clipboard in one command.

Writes your Claude usage limits to ~/.claude/usage-reporter/usage.json so other tools can read them

Sends each local session's status (what Claude is doing, task progress, context and usage) to the Claude Widget desktop app on this computer, and takes its Stop button

Paints Markdown onto Claude Code's prompt box as you type: fenced code becomes a syntax-highlighted card (even before you close the fence), and inline code, bold, italic, lists, task lists, quotes, tables, links and headings are styled. Paint only, your text is never changed.

Routes each turn and subagent to a model tier and reasoning effort using a classifier behind any server that speaks POST /v1/classify (model-router-api on Cloudflare Workers AI today), falling back to the engine's built-in classifier.

Adds /compact micro: elides Read results and strips thinking, keeping the conversation structure

/mejora y una frase: Claude te hace 3-5 preguntas, escribe un prompt con contexto (sin rol), lo puntúa y lo corrige hasta el 9/10, con bucle o sin bucle. Método de diariovida.com.

Subagente e agente de script (Workflow) criados sem modelo escrito rodam no Sonnet em vez de herdar o Opus; Opus só quando pedido de propósito

Ring Lucy's doorbell from Claude Code: /lucy-ping [send | retry | status [eventId]] through the Lucy bridge.

A decision model for other mods ($.decision.ask): typed questions about a state, answered by any System One endpoint (Jev by default) or a small Claude model.

A persistent JavaScript kernel the model drives with one eval tool: state survives between cells, and cells call Claude Code tools, models and subagents.

Restores the dim prompt suggestion after running a slash command like /copy

Loads skills from the cross-harness .agents/skills directories (project and ~) into Claude Code: listed for the model and runnable as / .

Replaces Claude Code's core system prompt with pi's coding agent prompt; /pi-prompt toggles it, remembered across sessions.

Keeps a short log of what you asked Claude to do and which files changed, across sessions, and writes your standup update with /standup.

Gives a session a random /color when it gets a new identity: the first /rename, unless the mod has already coloured it or seen you run /color, and every /branch.

Names and colours a session from metadata in the SKILL.md of a skill typed as / .