Commands

Shows each subagent of the session with its turns, time, model and tokens, the costliest first, yellow while it runs and green once it answered, and marks one that passed the token limit.

Submits a continue prompt while the task list has unfinished tasks, at most 99 times in a row by default, and turns the task tools on for every model unless CLAUDE_CODE_ENABLE_TODO_TOOLS is set.

Stops the model from repeating a tool call that just failed with the same input, until a file or command has changed something.

After a curl or wget an automated-client filter answered 403 or 429, gives the model the browser User-Agent to retry with, and the two cases where it must not.

Draw diagrams, LaTeX math, and tool-result images inline in kitty-graphics terminals

.NET development plugin for Claude Code — Roslyn semantic analysis, DI/architecture/EF Core guards, convention-aware scaffolding, TDD agents, security and performance review, and project lifecycle commands.

A band above the prompt showing your 5-hour and 7-day limits, context window and prompt-cache hit rate, styled for the terminal and the desktop app

A band above the prompt showing your 5-hour and 7-day limits, context window and prompt-cache hit rate, styled for the terminal and the desktop app

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.

A pixel Clawd in a side pane that reacts to what Claude is doing

Le flash info du dev au-dessus du prompt : les actus fraîches de Human Coders News, du Journal du hacker, de la veille de Camille Roux, de LinuxFr et des blogs IA (Anthropic, claude.dev, Claude Code…), une à la fois.

Specialists execution runtime in Claude Code: specialist activation and dispatch over MCP, session continuity hooks, and the supervising-activations skill. Consumes the Substrate work authority; it is not Substrate.

An output style that classifies each message by exchange type and shapes the reply to fit it: the answer first, a ceiling per type, reference codes that survive the session, and decisions handed back as questions

The client for ultrapowers: /ultrapowers commits an approved plan and runs on an exe.dev fleet you provision — a disposable sandbox runs the plan as a pool of tasks, a pool of builders, the plan's own probes and checks as the proof, and opens its own PR. No local engine.

Lists the files this session edited in a pane (/changes), with line counts and who edited them, beside the git working tree

Pins standing notes with /pin that Claude keeps following after /compact and /clear; shows them in one row above the prompt or as a counter, and keeps them per folder with --keep

Shows a live session pane (/ctx): context breakdown, usage limits with pace, tool calls and model requests; sends the model a one-time note when limits or context cross a threshold

Draws the mermaid diagrams of each answer as images in a pane (/show-me); needs mmdc and a kitty-graphics terminal such as Ghostty

Draws the current turn as a timeline in a pane (/timeline): model requests and tool calls per loop, where the time went, and the slowest steps

Apple Music now-playing band above the prompt: pixel album art, title, artist and a bopping Claude plush

Shows the repository's TODOs in a band above the prompt, newest first: TODO.md items and TODO/FIXME/HACK code comments dated with git blame, a Mine filter, issue links, and a toast when an edit adds one

Plan at the frontiers and review the plan in the browser: decisions and interfaces before mechanics, vertical slices each closed by a check. /vellum:start opens a planning mode of its own: Claude writes the plan and its mockups and grills the reviewer on the open choices; the reviewer comments, edits or approves in a page, and the answer reaches Claude as a prompt.

Benjamin's collection of Claude Code skills — daily workflow, code quality, dev tools, and specialist agents.

Routes each turn to a Claude model chosen by TypeSafe's Jev, asked through the Vercel AI Gateway.