Activepieces
Open-source AI workflow automation — 400+ MCP servers for AI agents
📋 Overview
Activepieces is an open-source workflow automation platform built from the ground up for AI agents — unlike n8n (which added AI later), AI is the core design principle. It provides a visual drag-and-drop builder with 400+ integrations and a unique MCP server marketplace where each integration is a callable tool for AI agents. The platform supports multi-agent orchestration, custom agent skills, and a growing library of pre-built AI workflows. MIT licensed with a generous free tier and self-hosting option.
✨ Key Features
- •23K+ GitHub stars — fastest-growing n8n alternative
- •400+ MCP servers as callable AI agent tools
- •Multi-agent orchestration with specialized agent roles
- •Pre-built AI workflow templates for common use cases
- •Visual drag-and-drop builder with custom code support
- •MIT licensed — no vendor lock-in, no usage caps
- •Self-hosted option with full data sovereignty
🎯 The Problem It Solves
Teams that want to build AI-driven workflows need more than a traditional automation tool with AI bolted on — they need native agent orchestration, MCP tool integration, and a marketplace of pre-built AI skills. Activepieces delivers this while remaining fully open-source.
🔧 How It Works
Workflows are built visually by chaining triggers (webhooks, schedules, MCP events), actions (API calls, AI agent invocations, data transforms), and logic (conditionals, loops, parallel branches). The AI agent nodes can call any of the 400+ MCP servers as tools, with built-in memory and context management. Multi-agent workflows coordinate specialized agents (researcher, writer, reviewer) through a central orchestrator. All workflows can be exported as JSON for version control.
🚀 Installation & Quick Start
Installation
docker run -d --name activepieces -p 8080:80 ghcr.io/activepieces/activepieces:latestQuick Start
- docker run activepieces
- Open localhost:8080
- Create a workflow
- Add AI agent node
✅ Pros
- •AI-first design — agents and MCP are core, not bolted-on
- •400+ MCP servers as callable tools for AI agents
- •Multi-agent orchestration with role-based coordination
- •Pre-built AI workflow templates for rapid deployment
- •MIT licensed — completely free self-hosted
- •Active community with weekly releases
- •Cloud free tier generous for small teams
❌ Cons
- •Smaller community than n8n — fewer tutorials and templates
- •Fewer non-AI integrations (400 vs n8n 400+, but n8n has more non-AI)
- •Visual builder less polished than n8n for complex workflows
- •Multi-agent features still maturing — occasional edge cases
- •Enterprise pricing requires sales conversation
💬 Practitioner Verdict
“Activepieces is the best open-source alternative to n8n for AI-first workflows — the MCP server marketplace is genuinely differentiated, and the multi-agent orchestration is more mature than n8n AI features. The trade-off: smaller community than n8n, fewer non-AI integrations, and the visual builder is less polished. For teams building AI agent workflows specifically, Activepieces is the better choice.”
Self-Hosted (Free)
Open source, MIT/Apache licensed. Run it yourself.
⭐ Star & Clone on GitHubFree forever. Your infrastructure, your data.
Activepieces Cloud Free
1K tasks/mo, 1 agent
- 1K tasks
- 1 agent
- Community support
Activepieces Cloud Pro
10K tasks, unlimited agents
- 10K tasks
- Unlimited agents
- Priority support
Activepieces Cloud Business
Unlimited tasks, SSO, audit
- Unlimited tasks
- SSO
- Audit logs
Enterprise
VPC, HIPAA, dedicated support
- VPC/HIPAA
- Dedicated support
- Custom SLA
Deployment Options
Ways to run Activepieces in production
📊 Specifications
- Language
- TypeScript
- License
- MIT
- Platform
- Linux, macOS, Windows
- Supported Models
- REST API, CLI
💰 Pricing Reality
Self-hosted is 100% free MIT-licensed. Cloud: Free tier (1K tasks/mo), Pro 5/mo (10K tasks), Business 0/mo (unlimited tasks, SSO, audit logs), Enterprise custom (VPC, HIPAA). AI API usage billed separately.