Commands

Push guard: before a git push to a public GitHub repo (or making one public) it scans what would go out for Supabase project IDs, keys, tokens, emails, local paths, private repo names and .env files, and asks in a one-key popup. /pushor toggles. Hungarian or English.

Snake in the side pane while Claude works: opens when you send a prompt, pauses when the turn ends, keeps your best score. /jatek be|ki. Hungarian or English.

Supabase guard: stops Supabase calls aimed at a project other than the current folder's, and asks in a one-key popup before destructive SQL, branch resets or project pauses. /supaor toggles. Hungarian or English.

Toolbox: a button at the bottom right of the prompt footer (or /eszkozok) opens switches above the prompt for Szabolcs's mods (focus mode, usage bar, project pane, snake game, Supabase guard, push guard, decision popup, ...). Hungarian or English.

Status line: model, context window, 5-hour and weekly limits with reset times, warning past 90%. Hungarian or English.

Passive 50-minute prompt-cache reminder with a keyboard/mouse panel: keep warm, ping once, or compact, with live cost estimates.

A live panel above your prompt: the plan as a stage bar, what Claude is doing in plain English, your asks ticked off, and a sound when it is done

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

Live pane of every ask you make in a session, ticked only with proof. Ask-splitter: right after you send a prompt, a band shows "I heard 3 asks" and numbers each one

Stops Claude reading .env and key files, masks API keys in command output, and lets you hand Claude a key through a pane without it ever seeing the value

Opens the newest page, image or video Claude made at the end of each turn, without stealing focus

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

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.

The best all-in-one mod pack for Claude Code: a usage limits and context HUD, a guard with undo for rm -rf and git reset --hard, .env and secret protection, and a receipt for every turn.

Keeps a long-running agent's working notes through every compaction: briefs the summarizer, re-attaches the notes after the summary, and shows them in a panel you can steer.

Warns when the context window passes a threshold (80% by default) and suggests /save-context before compaction

Sound packs for the moments that need you: a prompt sent, a permission prompt, a long job done, a subagent back, a failed test, a git push, a compaction. Packs from StarCraft, Warcraft, Age of Empires, Command & Conquer, Diablo, Counter-Strike, Duke Nukem, Rick and Morty, The Big Lebowski and more; add your own

Remembers which session a fork came from (renames never break it), shows it above the prompt, and sends a fork's report to its parent only when you approve it

GlanceFlow: see what Claude is doing, at a glance. Hides tool rows and shows one plain-English checklist above the prompt, with progress, helpers, history, a team report, pause and continue, and a password guard.

Side pane listing project, user and plugin agents with a run button each; /agents-panel toggles it.

A live side pane that maps a long Claude Code session: goal, current path, possible detours, decisions, open questions, checkpoints, working files and what to resume next. It observes on its own and asks before changing your intent.

Pass tasks between named Claude Code sessions: the receiver queues them, works each to a pull request, reports back and picks up the next. A band and panel show your sessions and tasks.

An opinionated engineering workflow for Claude Code: brainstorm -> plan -> TDD execution with fresh-context subagents -> verification -> captured learnings. Lean reviewers, stack-specific patterns for TypeScript/React, Python and Flutter, and six light safety hooks.

Keeps long-running Claude Code sessions alive: a usage band above the prompt, prompt-cache keep-alive across idle gaps, and a graceful pause at 95% of the 5-hour limit that wakes the session when it resets.