Weavr Documentation

Self-hosted workflow automation with AI agents

Weavr is an open-source workflow automation platform that combines the power of traditional automation tools with native AI agent capabilities. Run it on your own infrastructure with full data sovereignty.

Open Source Self-Hosted AI-Native Node.js 22+

Quick Start

Get up and running in under a minute.

# Clone and install
git clone https://github.com/openweavr/Openweavr.git
cd Openweavr
npm install

# Build backend and web UI
npm run build
npm run build:web

# Start the server
node weavr.mjs serve

Open http://localhost:3847 to access the web interface.

Key Features

AI Agents

Build workflows powered by Claude, GPT, and other LLMs with tool use, web search, and memory.

Event-Driven Triggers

React to GitHub events, cron schedules, webhooks, Telegram messages, and more.

MCP Support

Extend agent capabilities with Model Context Protocol servers for filesystem, databases, and APIs.

Full Control

Self-host on your infrastructure. Your data never leaves your servers.

Documentation

Getting Started

Installation, configuration, and your first workflow.

CLI Reference

Command-line interface for workflow management and execution.

Web Interface

Visual workflow management, execution monitoring, and chat.

Workflow Reference

YAML syntax, actions, triggers, and variable interpolation.

Integrations

GitHub, Telegram, Slack, Discord, Linear, and more.

AI Agents

Configure and use AI agents with tools, web search, and MCP.

Example Workflow

Build workflows visually or write YAML directly — both produce the same result.

Visual workflow builder with drag-and-drop steps

YAML Definition

The same workflow as YAML — schedule an AI research agent that sends daily market briefings to Telegram.

name: daily-market-research-telegram

trigger:
  type: cron.schedule
  with:
    cron: "0 9 * * *"

steps:
  - id: research
    action: ai.agent
    with:
      task: "Research today's top market insights..."

  - id: summarize
    action: ai.agent
    needs: [research]
    with:
      task: "Create a concise summary from: {{ steps.research.result }}"

  - id: send_to_telegram
    action: telegram.send
    needs: [summarize]
    with:
      chatId: "YOUR_CHAT_ID"
      text: "{{ steps.summarize.result }}"