DeckSpace is the command deck for agentic coding. One subscription for the whole deck — or own any tool outright. Bring your own key. $0 metering. Your code never leaves your machine.
› worktree ready — pick a task below› awaiting taskMost builders stitch a coding agent, a voice tool, a QA runner and an editor together by hand. DeckSpace ships them as one workspace — the same agents, memory and controls everywhere. No per-tool subscriptions. No credits. Bring your own key.
DeckSpaceLiveRun up to 16 AI coding agents at once — each isolated on its own git worktree — and command the whole deck by voice: Atlas opens rooms, spawns agents, runs missions and tests the result in a real browser.
Hold a key, speak, release — your words land at the cursor in any macOS app. Dictate prompts, commits and docs, or flip on a talk-back assistant. Text appears in under a second.
Point DeckTest at your site or app and it drives a real browser through your own Claude over MCP — it signs up, reads the verification email, pulls the code, clicks through, and reports what broke. Like a human tester, on demand.
Capture a region, a window or the whole screen — region capture on a global hotkey. Annotate it, and on-device OCR reads every word — then your clipboard holds one prompt with the image's absolute path, your marks and the recognised text, ready to paste into an agent.
Turn a phone clip and a screen recording into a finished, captioned tutorial for YouTube and TikTok — teleprompter on your iPad, auto-synced captions, one-tap render. The last mile of shipping: showing the world what you built.
Run several Claude accounts side by side — no log-out dance. A tiny native switcher for people who burn through one account's limit on Fable 5 or Opus, or keep work and personal apart.
These live inside DeckSpace and share one memory, one board and one set of controls — so every tool sees the same context.
Hand it a mission and it designs, ships and fixes in a loop — rewriting its own playbook to get sharper each pass. You set the goal; it opens the PRs.
Decompose one goal into a coordinated team — lead, scout, builders and reviewers — each on its own worktree, merging behind a review gate.
A living knowledge graph in plain markdown — [[wikilinks]] and all — shared with every agent over MCP. Context compounds session to session.
Two self-hostable MCP servers hand your memory graph and your task board to any agent tool that speaks MCP.
DeckSpace launches the CLI agents you already have installed and authenticated — mix and match models across panes. No new accounts, no lock-in. Bring your own key; DeckSpace never sees or stores it.
Start from a Kanban card, a repo, or a rough idea. The session becomes a focused deck instead of a pile of tabs.
Terminal panes, file tree, editor and agent launch points come up together — one place for the work.
Spin up one agent or decompose a goal into a parallel crew on isolated worktrees, and watch status change live.
Jump between code, shell output and the diff. A verification gate runs tests, lint and a secret scan before you merge.
DeckSpace is a native desktop app (Tauri + Rust) — small, fast and offline-capable. The terminal, editor, board and memory all run locally. The only network calls are the ones your own agent CLIs make to your own model provider.
Use the CLIs you already pay for. DeckSpace never meters or marks up your model usage — you pay your provider directly. DeckSpace itself is a flat subscription, never per-token.
Confine an agent's writes to the repo, reroute destructive commands to a recoverable trash, and snapshot before every action.
Block cloud-metadata endpoints and SSRF. Telemetry is off by default — nothing about your code goes anywhere.
Every plan starts with a 7-day free trial — card required, cancel anytime. Flat fees, bring-your-own-key, $0 metering. Prefer to own a tool outright? See the one-time apps.
Or own any tool outright — DeckVoice $19 · DeckTest $29 · Claude Accounts $14, bring your own key, $0 metering. Complete licenses deactivate if you cancel; owned licenses never do.
A developer ecosystem for agentic coding. At the center is a native desktop "command deck" for running AI coding agents in parallel — a grid of agent panes, a Kanban board, a swarm, diff review, a memory graph, autonomous loops and voice. Around it sit DeckVoice (voice), DeckTest (browser QA) and DeckStudio (tutorial studio), all sharing the same agents and memory.
A native desktop command deck that runs the agent CLIs you already installed — up to sixteen at once, each in its own real terminal on its own git worktree — with a board that feeds them work, a review gate before anything merges, a memory graph they all share, and a voice assistant that can drive the whole app.
Twelve CLI agents today: Claude Code, Codex, Aider, Antigravity, OpenCode, Cursor Agent, Droid, Copilot CLI, Grok CLI, Amp, Qwen Code, and a legacy Gemini entry. You can also register any other CLI yourself. DeckSpace detects whichever are on your PATH.
No. DeckSpace is bring-your-own-key — you authenticate the agent CLIs once, the way you already do, and DeckSpace never meters, marks up or brokers your model usage. Pro is $24/mo (or $228/yr) for DeckSpace itself. Deck Complete is $39/mo (or $372/yr) and adds DeckVoice, DeckTest and Claude Accounts while you subscribe. Deck Complete Team is $69 per seat per month for 2 to 50 seats. Every plan starts with a 7-day free trial.
No. The terminal, editor, board and memory all run locally and offline. The only outbound calls are the ones your own agent CLIs make to your own model provider. Telemetry is off by default.
macOS 12 and newer. It is a native Tauri + Rust app. Every Deck app is macOS today, except Claude Accounts, which also ships a Windows launcher.
DeckSpace for macOS is live — start your 7-day free trial and cancel anytime.
Want DeckSpace on Windows or Linux, or early access to DeckStudio? Tell us where to reach you — we'll be in touch fast.
Thanks — we'll email you your build and access details shortly.