OpenClaw
Open-source local-first AI agent that runs on your device and handles real tasks — email, calendar, browser, files — controlled via WhatsApp, Telegram, or iMessage.
Overview
OpenClaw is an open-source AI agent that lives on your machine and does real work on your behalf. Think of it as a local assistant that can read and send emails, manage your calendar, control your browser, manipulate files, and execute shell commands — all triggered by a simple message from your phone via WhatsApp, Telegram, or iMessage.
What sets OpenClaw apart from cloud-based automation tools is its local-first architecture. Everything runs on your hardware, which means your data never leaves your device and you're not paying for someone else's compute. You bring your own LLM API keys (OpenAI, Anthropic, or others), so you pick the model that fits your budget and quality needs. The project grew out of the earlier Clawdbot experiment and exploded on GitHub and X in early 2026 with users sharing stories of agents autonomously running business operations and building websites.
The trade-off is clear: this is a tool for people comfortable with self-hosting. You'll need to install dependencies, configure API keys, and troubleshoot issues yourself. There's no polished web dashboard or managed service — it's a developer-oriented project with a growing but still maturing ecosystem.
Key features
Local AI Agent
Runs entirely on your device with no cloud dependency. Handles email, calendar, browser automation, file management, and shell commands autonomously.
Messaging Control
Send tasks to your agent via WhatsApp, Telegram, or iMessage. Get status updates and results back in the same conversation thread.
Task Automation
Chain multi-step workflows like researching a topic, drafting a document, and emailing it — all from a single message. Uses your own LLM API keys for flexibility.
Pricing
Free tier: Completely free and open-source — you only pay for your own LLM API usage (e.g. OpenAI or Anthropic keys)
| Plan | Price | What's included |
|---|---|---|
| Self-Hosted | Free | Full feature set, open-source, unlimited usage — bring your own LLM API keys |
Full feature set, open-source, unlimited usage — bring your own LLM API keys
Pros & cons
Pros
- ✓Fully free and open-source with no usage limits or subscription fees
- ✓Local-first architecture keeps all data on your device — strong privacy story
- ✓Messaging-app interface means you can control your agent from anywhere on your phone
- ✓Bring-your-own-model approach avoids vendor lock-in and lets you optimize cost vs. quality
Cons
- ×Requires technical setup — installing dependencies, configuring API keys, and self-hosting
- ×No managed cloud option or polished GUI — terminal and config files only
- ×LLM API costs still apply and can add up with heavy autonomous use
- ×Young project with a fast-moving codebase — expect breaking changes and rough edges