AI AGENTS
SwarmClaw
by SwarmClaw
Overview
Open-source, self-hosted AI agent runtime and multi-agent framework for autonomous agent swarms with memory, delegation, schedules, and coding-agent integrations.
Details
SwarmClaw describes itself as an open-source, self-hosted AI agent runtime and multi-agent framework for autonomous agent swarms. Its documentation positions it as a runtime and control plane for running agents, connecting providers, delegating work, automating recurring work, managing context, and bridging agents into external systems. Supplied docs also mention memory, MCP tools, schedules, delegation, coding-agent integrations, and support for 25 built-in providers plus custom OpenAI-compatible endpoints.
When to Use
Use SwarmClaw when evaluating a self-hosted runtime and control plane for AI agents. Use it for multi-agent swarm workflows that need memory delegation recurring scheduled work or external-system bridges. Use it when you need to connect agents to multiple model/provider options including built-in providers and custom OpenAI-compatible endpoints.
Getting Started
- Read the introduction in the SwarmClaw docs: https://www.swarmclaw.ai/docs
- Follow the getting-started guide for prerequisites and global npm installation via @swarmclawai/swarmclaw.
- Launch SwarmClaw for the first time
- complete the setup wizard
- and access the local server as described in the getting-started documentation.
- Review provider setup in the providers docs if you need built-in providers
- local runtimes
- CLI providers
- gateways
- or custom OpenAI-compatible endpoints.
Key Features
- •Self-hosted AI agent runtime and control plane
- •Multi-agent framework for autonomous agent swarms
- •Agent memory and context management
- •Delegation and recurring scheduled work
- •MCP tools and external-system bridging
- •25 built-in providers plus custom OpenAI-compatible endpoints
- •Support for CLI providers
- •API providers
- •local runtimes
- •OpenClaw gateways
- •Goose
- •Hermes Agent
- •and model failover
Capabilities
- •agent-runtime
- •multi-agent-orchestration
- •self-hosting
- •agent-memory
- •task-delegation
- •scheduled-automation
- •mcp-tools
- •provider-routing
- •openai-compatible-endpoints
- •local-runtime-support
- •model-failover
Last updated Jun 11, 2026