Frequently asked
questions

Yes. ClawHQ is MIT licensed and the full source is on GitHub. The entire platform — agents, channels, runtime, dashboard — is free to self-host with no per-seat fees.

Teams and individuals who want a full AI workforce operational immediately — without building agents from scratch, wiring up integrations, or managing a complex LLM stack. Also teams with privacy requirements who can't send data to cloud AI platforms.

Clone the repo, configure your .env file with your LLM API key and channel tokens, then run docker compose up -d. The full stack comes up in one command. Full walkthrough: clawhqplatform.com/docs/quickstart.

Any machine that runs Docker. For production use, 2 vCPU and 4GB RAM is a reasonable baseline. ClawHQ runs on Linux, macOS, and Windows with Docker Desktop. Works on VPS, bare metal, local machine, and cloud VMs.

Yes. ClawHQ is air-gap capable and suitable for isolated networks, compliance environments, and regulated industries. Pair with a local LLM provider (Ollama) to remove all external API dependencies.

Any OpenAI-compatible API: Anthropic (Claude), OpenAI (GPT-4o), Gemini, DeepSeek, Groq, Together AI, and local models via Ollama. Set your provider in .env — all agents use it automatically.

30+ agents active from day one: Researcher, Email Assistant, Doc Writer, Coder, Analyst, Social Media Manager, Planner, Translator, Legal Assistant, Sales Assistant, Security Auditor, Recruiter, Customer Support, Test Engineer, Architect, Tutor, and more. No configuration required — they're live immediately after deployment.

Discord, Slack, Telegram, and Webhook. All channels are pre-configured in the stack — provide your bot tokens in .env and they connect automatically. No custom integration code required.

Yes. Custom agents and skill packs can be installed via ClawMart, the community pack registry. You can also author your own packs and load them into your ClawHQ instance.

The Orchestration layer routes tasks to the right agent based on intent, and agents can delegate sub-tasks to specialists. For example, a Researcher agent can hand off a draft to the Doc Writer, which hands off to the Email Assistant for delivery — without manual intervention.

Six pre-configured services: OpenClaw (agent gateway — routes messages from channels to agents), OpenFang (agent runtime built in Rust — sandboxed, autonomous, fast), Paperclip (per-agent memory and context), Hermes (context compression — reduces token usage), Mem0 (long-term memory), and an Orchestration layer for multi-agent coordination. All wired together. You don't configure the connections — they're done.

OpenFang is the autonomous agent runtime underneath ClawHQ, built in Rust. Each agent runs sandboxed with full tool access — web search, code execution, GitHub, file system, and more. The Rust runtime means low memory overhead and fast parallel execution across many agents.

Dify is a visual LLM app builder — you construct agents and workflows from scratch using a drag-and-drop editor. ClawHQ ships with 30+ agents already deployed and coordinating. You don't build — you delegate. ClawHQ is also channel-native (Discord, Slack, Telegram pre-wired); Dify requires custom integration for each channel. Full comparison →

Flowise is a drag-and-drop LangChain UI for building individual AI pipelines. ClawHQ is a complete team of pre-deployed coordinating agents. Flowise gives you the tools to build; ClawHQ gives you the team. Full comparison →

n8n is a workflow automation platform — you build trigger-based automations using a visual editor. ClawHQ agents are autonomous: they reason, plan, use tools, and handle ambiguity without a predefined workflow. They're complementary: n8n for deterministic automations, ClawHQ for tasks that require judgment. Full comparison →

No. ClawHQ is self-hosted with zero telemetry — no analytics, no usage reporting, no callbacks to any external server. All data stays within your own infrastructure. The only external connections are the LLM API calls you configure (which go directly to your chosen provider, not through us).

Yes. ClawHQ is air-gap capable — deployable in fully isolated networks with no external connectivity. Pair with a local LLM provider (Ollama) to eliminate all external API calls entirely. Suitable for defense, healthcare, legal, and financial compliance environments.