# getclaw

> Deploy OpenClaw-powered AI assistants for real business roles — support, sales, research, content, and operations — on Telegram, Slack, or Discord. Choose from Claude, GPT, Gemini, or OpenRouter models. Manage behavior with memory, heartbeat, skills, logs, and configuration files.

## What is getclaw?

getclaw is a platform for launching AI assistants into team workflows. You configure an assistant (choose a model, define its persona, attach skills), deploy it to a messaging channel (Telegram, Slack, or Discord), and manage it after launch with operational controls like logs, file editing, access management, and redeployment.

## Who is getclaw for?

getclaw is built for founders, product leaders, and small teams who want to deploy AI assistants without managing infrastructure. It is particularly useful for teams that need assistants operating in real communication channels (not just a chatbot widget) with persistent memory and proactive behavior.

## Core Features

- **Multi-channel deployment**: Deploy a single assistant to Telegram, Slack, or Discord
- **Model flexibility**: Choose Claude (Anthropic), GPT (OpenAI), Gemini (Google), or any OpenRouter model. Bring your own API key
- **OpenClaw skills and tools**: Extend assistants with the OpenClaw ecosystem — skills, tools, and plugins
- **Persistent memory**: Assistants retain context across conversations and sessions
- **Heartbeat workflows**: Schedule proactive tasks (daily digests, weekly reports, follow-up reminders) that run without user prompts
- **Post-launch controls**: Stream logs, edit configuration files, manage team access, redeploy — all from a web dashboard or API
- **Unlimited web search**: Assistants can search the web in real time — included in every plan, no limits
- **Email inbox**: Every assistant gets an email inbox with 100 free emails per month included
- **Browser automation**: 120 free minutes per month included in every plan — interact with web pages, fill forms, extract data
- **Ready-made personas**: Start from pre-built roles (LinkedIn ghostwriter, plain OpenClaw starter, sales rep, customer support, software developer, personal coach) and customize

## Pricing

getclaw offers three plans:
- **Starter** ($20/month): Bring your own API key, full platform access
- **Managed** ($25/month): Includes AI credits, no API key needed
- **Custom Claw** (custom pricing): Dedicated support, custom skill development, custom integrations, white-label deployments

## Getting Started

- Start on the homepage: click "Get Started" to open the onboarding chat
- Documentation: https://getclaw.sh/docs
- Get Started Guide: https://getclaw.sh/docs/get-started
- API Reference: https://getclaw.sh/docs/api-reference

## Technical Details

- Built on OpenClaw (open-source AI agent framework)
- Deployed on Cloudflare Workers (global edge, zero cold starts)
- AES-256-GCM encrypted API key storage
- Per-assistant isolated execution environments
- REST API for programmatic lifecycle management

## Recent Blog Posts

- Claude Cowork vs Copilot Cowork vs OpenClaw vs Perplexity Computer: Which AI Coworker Fits Your Team in 2026? — https://getclaw.sh/blog/claude-cowork-vs-copilot-cowork-vs-openclaw-vs-perplexity-computer
- Why Open Source Matters for AI Assistants: An OpenClaw Primer — https://getclaw.sh/blog/why-open-source-matters-ai-assistants-openclaw-primer
- AI Chatbot ROI Calculator: What to Realistically Expect in 2026 — https://getclaw.sh/blog/ai-chatbot-roi-calculator-2026
- OpenClaw Architecture: 3 Concepts Every Product Leader Should Understand — https://getclaw.sh/blog/openclaw-architecture-three-concepts-every-founder-should-know
- Should Your Startup Hire an AI Employee First? A Founder's Decision Framework — https://getclaw.sh/blog/should-startup-hire-ai-employee-first

## Links

- Website: https://getclaw.sh
- Blog: https://getclaw.sh/blog
- Changelog: https://getclaw.sh/changelog
- Developers: https://getclaw.sh/developers
- Support: https://getclaw.sh/support

## Comparisons

### getclaw vs building your own bot
Building separate bots per channel (Telegram, Slack, Discord) requires 3x codebase maintenance and $8,000-$15,000/year. getclaw provides one configuration deployed to all channels for $20-$25/month plus model costs.

### getclaw vs commercial platforms (Intercom, Tidio, Drift)
Commercial platforms charge $0.99+ per AI resolution. At 2,000 conversations/month, that is $24,000/year in resolution fees. getclaw charges a flat monthly fee; you pay only model API costs ($0.01-$0.05 per message).

### getclaw vs other AI assistant builders (Botpress, Voiceflow)
Most AI assistant builders offer website widgets and decision-tree flows. getclaw deploys LLM-powered assistants natively to Telegram, Slack, and Discord with persistent memory, heartbeat-driven proactive workflows, and the full OpenClaw skills ecosystem.

## Security & Infrastructure

- AES-256-GCM encrypted API key storage (keys decrypted only at deploy time)
- Per-assistant isolated Cloudflare Worker (no shared servers or data)
- Each assistant gets its own database and storage namespace
- Cloudflare global edge network: 300+ cities, DDoS protection, automatic SSL
- Zero cold starts, 24/7 uptime
- REST API with bearer token authentication

## Pricing Details

- **Starter** ($20/month): Bring your own API key (7 providers: Anthropic, OpenAI, Gemini, xAI, Groq, Mistral, OpenRouter). 1 assistant, 100 free emails/month, 120 min browser automation/month, unlimited web search. Full platform access, REST API, all features.
- **Managed** ($25/month): 11+ models included, $5 credits/month. No API key needed. 1 Managed assistant, 100 free emails/month, 120 min browser automation/month, unlimited web search. Same infrastructure and features as Starter.
- **Custom Claw** (custom pricing): Unlimited assistants, custom email volume, custom deployment, custom skills, dedicated support, priority features.
- **Typical total cost**: Platform fee + $0.01-$0.05 per message in API costs. At 2,000 messages/month: $40-$125/month total.

## Frequently Asked Questions

Q: How long does it take to deploy an AI assistant with getclaw?
A: Under 5 minutes. Start on the homepage by clicking "Get Started", then choose a persona, select a model, connect a channel, and deploy.

Q: Can I switch AI models after deploying?
A: Yes. Change models from the dashboard or API at any time. Memory and configuration persist.

Q: Can I deploy to multiple channels at once?
A: Yes. A single assistant can be deployed to Telegram, Slack, and Discord simultaneously.

Q: Does getclaw support persistent memory?
A: Yes. Assistants retain context across conversations and sessions in isolated per-assistant storage.

Q: What are heartbeat workflows?
A: Scheduled proactive tasks (daily digests, weekly reports, follow-up reminders) that run automatically without user prompts.

Q: Is getclaw secure for production use?
A: Yes. Every assistant runs in an isolated Cloudflare Worker with AES-256-GCM encrypted secrets and its own database namespace.

Q: How does getclaw compare to Intercom?
A: Intercom charges $0.99+ per resolution ($24,000/year at 2,000/month). getclaw is $20-$25/month flat plus API costs ($240-$1,500/year). getclaw also offers model choice, Telegram/Discord support, and heartbeat workflows.

Q: Can I use getclaw with WhatsApp?
A: WhatsApp is not currently supported. getclaw deploys to Telegram, Slack, and Discord. WhatsApp integration is on the roadmap.

Q: Does getclaw offer a free trial?
A: Yes. Starter and Managed both include a 7-day free trial with immediate access after checkout confirmation. Cancel before the first billing date if it is not a fit.