check_circle
Foundation v1.1 — Now Available

Your Personal AI Agent.
Zero Code Required.

Design, personalize, and deploy your own intelligent core for free today. Simple enough for anyone, powerful enough for everyone.

See How It Works arrow_downward
$0 To Start
20+ AI Skills
6+ AI Providers
Built for researchers, developers, founders & everyday power users
schoolResearchers
codeDevelopers
rocket_launchFounders
personEveryone
How It Works

Idea to Agent
in 60 Seconds.

01

Connect Your Sources

Point to your files, socials, or websites. Biome ingests and maps your data automatically — no technical setup required.

02

Configure Your Agent

Drag and drop capabilities. Set your agent's personality, skills, and triggers through our visual no-code interface.

03

Ready in One Click

Hit deploy and your agent is live on your Mac. No cloud account, no server costs — everything runs locally on your machine.

$ biome deploy

Connecting your data...        [Done]
Loading AI skills...           [Done]
Starting browser relay...      [Done]
Running agent health check...  [Done]

✓ Agent is live.
 Running locally on your Mac
Biome v1.1.0
Key Features

A Dedicated Intelligence Core.
Just for You.

schedule

Deep Context Awareness

Biome remembers every conversation and builds persistent long-term memory. The more you use it, the smarter and more personal it becomes.

hub

20+ AI Skills

Web search, live code execution, file and image attachments, and browser automation — all built in. Attach PDFs, images, and code files directly to any message.

shield

Private by Default

Your conversations, memory, and API keys are stored locally on your device — never on our servers. API calls go directly from your Mac to your chosen AI provider, not through us. Granular permission modes let you control exactly what the AI can touch.

account_tree

Visual Logic Builder Beta

Build complex multi-step workflows without a single line of code. Branch logic, conditionals, and loops — all visual, all yours.

model_training

Bring Your Own Model

Use Claude, GPT-4, Gemini, or any OpenAI-compatible endpoint. Switch models per workflow step without rebuilding anything.

payments

Start for Free. Stay Free.

The free tier is permanent, not a trial. We make money when you scale, not when you're getting started.

attach_file

File & Image Attachments

Attach PDFs, images, code files, CSVs, and more to any message. The AI reads and reasons over your files directly — no upload to a third-party server.

history

Full Conversation History

Every session is saved locally and instantly searchable. Pick up any previous conversation exactly where you left off — no cloud sync required.

play_circle

Chat-Triggered Workflows

Build and name your workflows once, then run them from anywhere — just type run workflow My Pipeline in chat and Biome executes it instantly.

Browser Control System Beta

Your AI lives in your browser.

Biome ships with a background browser relay that gives the AI direct, real-time control over your Chrome browser — open tabs, navigate pages, click elements, read content, and automate anything you can do manually. No cloud infrastructure. No screen sharing. Completely local.

tab Open & manage tabs
travel_explore Navigate to any URL
ads_click Click & type on pages
article Read page content
push_pin Pin & sort tabs
screenshot_monitor Screenshot & inspect

Powered by a lightweight local relay server (port 9999) + a Chrome extension. Enable in Settings → General → Browser Control. The app starts the server automatically.

extension Get the Chrome Extension

Required for browser automation workflows.

open_in_browser
tab Tab 1
tab Tab 2
Live browser control
Browser Control → AI Chat Piggyback

Use any AI chat in your browser — no extra API key.

Because Biome already controls your browser, it can automatically relay queries to whichever AI chat you have open and pipe the response back. Works with any AI chat interface in Chrome. Your own session, your own subscription.

auto_awesome

Claude

Recommended

Biome works best with an active claude.ai session open in Chrome. Claude's longer context window and instruction-following make it ideal for complex agentic tasks via browser piggyback.

chat

Any AI chat in your browser

The piggyback system is not Claude-specific — it works with ChatGPT, Gemini, Perplexity, and any other AI chat interface you have open in Chrome. If you can type into it, Biome can use it.

* A primary model (API key or Biome plan) is still required for Biome's core reasoning and tool loop. The AI chat piggyback is an optional enhancement layer for web-augmented responses.

Pricing Plans

No hidden fees. No trial periods. Start building today.

01 / Personal
Free Forever
$0 USD — No card required
  • checkFull chat, mindmap & skills
  • checkBrowser Control system
  • checkAll AI models (bring your own key)
  • check500 long-term memory entries
  • checkCommunity support
02 / Professional Most Popular
$29/mo
Billed Monthly · Cancel Anytime
  • checkEverything in Free
  • checkExtended memory (2,500 entries)
  • checkZero-Knowledge privacy mode
  • checkPriority email support
  • checkEarly access to new features
03 / Business
Custom
Let's Talk
  • checkEverything in Pro
  • checkShared team workspace Soon
  • checkCustom privacy controls
  • checkWhite-glove onboarding
Contact Us

Start Building.

Download the Biome desktop app and deploy your first personal AI agent in minutes. Always-on intelligence awaits.

shield macOS Security Tip — If your Mac blocks the app on first launch, simply go to your Mac's System Settings > Privacy & Security, scroll down, and click Open Anyway. This is a normal one-time confirmation for independently downloaded software. Learn more →

Simple Support

Our team and community are here to help you get the most out of your personal AI. No jargon, just clear solutions.

openclaw.solutions@gmail.com arrow_forward

Average response time: 2 hours. We're a small team that actually reads every message. No bots, no ticket queues on the Personal plan.

Yes — Biome runs entirely on your machine and calls AI APIs directly using your own keys. You bring keys for whichever providers you want: Google Gemini (free tier available), Anthropic Claude, OpenAI, xAI Grok, Mistral, or Perplexity. Keys are stored locally, never sent to our servers. Gemini has a generous free tier, so you can get started at zero cost.

Browser Control is an optional feature that gives the AI direct access to your Chrome browser — it can open tabs, navigate pages, click elements, read content, and more. It requires a lightweight Chrome extension (loaded from the Biome app in one click) and a local relay server that Biome starts automatically. The core chat and agent features work fine without it.

Biome supports Google Gemini (2.0 Flash, 2.5 Flash, 2.5 Pro), Anthropic Claude (3.5 Sonnet, 3.5 Haiku, 3 Opus), OpenAI (GPT-4o, GPT-4o mini, o1), xAI Grok-2, Mistral (Large, Nemo), and Perplexity Sonar. Add an API key for any provider in Settings → API Keys and those models instantly appear in the model picker.

Yes — through OpenClaw, Biome's AI skill system. Built-in skills include web search, live code execution, file and image attachments, browser automation, and multi-step workflow pipelines. You can also add custom data source URLs in Settings that the AI will reference in every conversation. If you need a specific integration, email us — we prioritise community requests.

No. The free download gives you the full Biome app — chat, mindmap, skills, browser control, all models — with 500 memory entries and community support. No trial clock, no feature gates, no upgrade nags. Pro adds extended memory, zero-knowledge privacy mode, and priority support for people who want more.

Almost everything stays on your machine. Conversations, memory, and settings are stored in localStorage — never on our servers. API calls go directly from your device to whichever AI provider you chose. We only see your email address (used for login). Pro users can enable Zero-Knowledge mode for full end-to-end privacy. We are GDPR and CCPA compliant.

Latest v1.1 — Visual Workflow Builder
v1.0.4 — Unlimited Data Sources
v1.1.0

Workflow Builder — Mar 2026

New visual logic builder with branching, conditionals, and filter steps. Build multi-step AI workflows entirely on your local machine.

v1.0.4

Unlimited Source Mapping — Feb 2026

Removed the 10-source cap on the free tier. All plans now support unlimited data source connections.

v1.0.2

Notion + Linear Integrations — Jan 2026

Native Notion database sync and Linear issue tracking. Agents can now create, update, and query both platforms directly.

v1.0.0

Foundation Release — Dec 2025

Initial public launch. Multi-model chat, long-term memory, browser control, and one-click deployment.