VoidSoul
Assistant
The Jarvis loop, finally local.
A floating desktop AI that talks, listens, sees your screen, drives your mouse, opens your apps, edits your files, and remembers every conversation. Bring whichever AI you already love — 12 providers, one body.
Builds aren't live yet — Founder's Edition launches when downloads go public.See pricing →
Live surface
See it, floating.
Real panels from the running app — chat with live agent steps, the multi-provider picker, the cost dashboard, the Raycast-style overlay. Hover the canvas; it tilts with you.
Chat panel
Active model
12 providersAnthropic
claude-sonnet-4-5
OpenAI
gpt-4o
Ollama
qwen2.5:14b
gemini-2.0-pro
LM Studio
llama-3.3-70b
Groq
llama-3.3-70b
Switch mid-conversation. All memory, projects and tools follow.
12 providers
May 2026
$18.60
Cost dashboard
Quick AI
Why it exists
The AI you already pay for —
with a body.
Today's AI apps split awkwardly. Cloud chatbots can't touch your computer. Editor copilots are trapped in one IDE. Closed ecosystems lock you to one vendor and one model. None of them remember you next month.
VoidSoul Assistant is the third path. A single persistent assistant that's always there, can do real work on your machine, speaks every major model, never sends your data anywhere you didn't tell it to, and grows with you through plugins, MCP servers, and per-mode memory.
Not a better chatbot — an operating layer for whatever AI you already pay for.
What you can do
Real workflows, not chat-toy demos.
- ✦Tell it to read your error log, fix the bug, run the test suite, and commit.
- ✦Speak a thought aloud — it transcribes, refines, and pastes into your editor.
- ✦Paste a screenshot — it identifies the failing widget and writes the patch.
- ✦Schedule it to grep your codebase for unused exports every morning.
- ✦Drop a PDF into the chat — talk to it without any "Pro plan" upsell.
- ✦Use Claude for reasoning, GPT-4o for vision, Ollama for cheap follow-ups, all in one thread.
- ✦Wake the orb with a phrase. Ask a question while doing something else. Hear it answered.
- ✦Chain MCP servers — read a Notion page, query GitHub, write a file, all in one prompt.
What's in the box
Nine pillars.
Four-layer memory
Threads, story-so-far summaries, durable facts, RAG over chat + indexed files. The model never forgets — and follows you across providers.
18 agent tools + MCP
Open apps. Run shell. Read & write files. Drive mouse and keyboard. OCR your screen. Search the web. Run Python. Plus any MCP server.
Vision + screen control
see_screen captures your screen as an image the model literally looks at. Combine with mouse control and the assistant uses your GUI.
Local voice loop
Whisper STT and the wake word run locally — no API key, no audio leaving your machine. Speak mid-reply to interrupt.
12 providers
Anthropic · OpenAI · Gemini · Ollama · LM Studio · llama.cpp · Groq · xAI · OpenRouter · DeepSeek · Mistral · Custom. Switch per-thread.
Raycast-style Quick AI
Cmd+Shift+J from anywhere. One-shot answers, reads your clipboard, no thread, no config. The fastest path from question to answer.
Projects & Notebooks
Pinned mode + custom instructions. Notebook cells for prompt, python, search, markdown. Schedule prompts on a cron.
Per-token cost only
No subscription floor. Bring your own keys, pay for what you use. Dashboard with daily-spend chart and budget alerts.
Local-first by default
Keys encrypted in your OS keychain. History in local SQLite. No telemetry. Pollinations, DuckDuckGo, and local Whisper as keyless defaults.
Full setup
Seven steps to unlock everything.
From first launch to wake-word voice, MCP plugins, and budget alerts. Tap a step to preview what setup actually looks like.
Step 01
Install & summon the orb
Download the installer for your OS. First launch drops a glowing orb on your desktop and a tray icon — that's the assistant.
Press Cmd/Ctrl+Shift+Space from anywhere to summon. Drag the orb to wherever it suits your workflow.
$ download → install → launch
▸ orb mounted
▸ tray icon active
▸ hotkey registered (⌘⇧Space)
✓ ready
Up close
The details
you'll feel.
Real UI surfaces from the running app. Hover for the small lift.
Floating orb
Always there, never in the way.
Drag-to-move. Click-to-summon. Tray-resident on every OS.
Live agent
See what it's doing, as it does it.
A live strip while the agent works — searching, reading, running, clicking — never silent.
Capped at six agent steps · partial output preserved on Stop · every step logged.
Quick AI
One shortcut. Every answer.
Press Cmd+Shift+J from anywhere — Word, Photoshop, your terminal. Ask. Done.
Cost honesty
Spend, daily. Budget, enforced.
Pay for tokens, not for a chat UI. Alerts at 75 / 90 / 100% of monthly cap.
May 2026
$18.60
Privacy & permissions
It never
acts silently.
Every capability that touches your machine is gated behind a permission you explicitly grant — and can revoke any time. File writes are reversible. Every action is logged. Private mode strips the chat: no save, no facts, no screen context.
Keys in your OS keychain
API keys never leave the main process. Encrypted by Windows DPAPI / macOS Keychain / Linux libsecret.
No telemetry, no cloud
We don't run a server. No usage tracking, no crash phone-home, no analytics SDKs.
Source-visible licence
The code lives on GitHub for transparency — anyone can audit what the app actually does. Resale is reserved.
Per-request abort
Hit Stop and the model call, the tool subprocess, and any background fetch cancel in the same breath.
Permissions, explicit
Terminal
Shell commands · sandboxed Python
Filesystem
Read, write, organise files (writes are undoable)
App control
Launch apps, foreground windows
Input access
Keyboard + mouse control
Screen capture
Screenshots, OCR, vision
Microphone
Voice input + wake word
Browser
Opening URLs
Every prompt that triggers a privileged action surfaces a dialog with the exact tool name and arguments. You always know what's about to run.
See the loop
90-second demo.
Wake word → question → multi-step agent run → spoken reply. The full Jarvis loop, end-to-end.
Devlog clips on YouTube meanwhile.
Pricing
One price. Yours forever.
Cancel ChatGPT Plus. Cancel Claude Pro. Pay once. Bring your own keys. No subscription floor, ever.
Free Forever
$0
Local-only. No commitment.
- ✓Ollama, LM Studio, llama.cpp
- ✓Five threads max
- ✓Local voice loop
- ✓Basic agent tools
Earliest adopters
FREE
$129
Full app · lifetime · zero cost
- ✦Everything in Founder's Edition
- ✦All 12 providers · MCP · voice loop
- ✦Direct line for feedback
- ✦Forever, no strings
First three downloads at launch. Then Founder's Edition kicks in.
Lifetime
$129
$199
AUD · one-time · yours forever
- ✦All 12 providers · MCP support
- ✦Full voice loop · wake word · barge-in
- ✦Projects · Notebooks · Quick AI
- ✦Scheduled tasks · cost dashboard
- ✦All themes & languages
- ✦Every future update, free
Climbs to $249 after the first 100.
All prices in AUD. Paid plans handle their own provider costs (you bring your own API keys). VoidSoul never proxies your model traffic.
Talk to the orb.
When the first build ships, you'll see it here. Until then, follow the studio devlog for the build-in-public ride.