Skip to content

neul-labs/m9m

Repository files navigation

m9m

Automate repetitive work with one command. No servers, no dependencies, no Node.js runtime. Just a 30MB binary that runs your workflows.

Connect apps, process data, schedule tasks, and run AI agents -- all from your terminal.

Build Status Go Report Card Coverage Go Reference Release License


What You Can Automate

Business Operations

  • Process new signups and route them to the right onboarding flow
  • Score leads based on activity across multiple platforms
  • Route invoices to approvers based on amount and department
  • Segment customers by behavior and sync to your CRM

Data Processing

  • Sync databases on a schedule with conflict resolution
  • Merge reports from 5 different services into one dashboard
  • Clean and validate incoming data before it hits your database
  • Transform CSVs, JSON, and API responses into the format you need

AI & Developer Workflows

  • Run Claude Code in a sandboxed environment to analyze codebases
  • Process customer feedback with GPT-4 and route by sentiment
  • Auto-create GitHub issues from support tickets
  • Chain multiple AI models together with human review steps

Scheduled Tasks

  • Back up critical data to S3 every night
  • Generate and email weekly reports automatically
  • Clean up expired records and notify stakeholders
  • Monitor APIs and alert your team on Slack when something breaks

Browse ready-to-use examples in the examples/ directory.


See It In Action

1. Install

curl -fsSL https://raw.githubusercontent.com/neul-labs/m9m/main/install.sh | bash

Or download from GitHub Releases. Docker also available: docker run -p 8080:8080 neul-labs/m9m:latest

2. Run the demo

m9m demo

You'll see six workflows execute in milliseconds:

m9m Capability Demo
===================

Demo 1: E-Commerce Order Processing
  Pipeline:  Start -> Set -> Set -> Filter
  Status:    PASS
  Duration:  8ms

Demo 2: Customer Segmentation
  Pipeline:  Start -> Set -> Filter -> Set
  Status:    PASS
  Duration:  12ms

Demo 3: Multi-Branch Routing
  Pipeline:  Start -> Set -> Switch -> Set(x3)
  Status:    PASS
  Duration:  8ms

...

Results: 6/6 demos passed

3. Run your own workflow

m9m exec workflow.json
m9m exec workflow.json --input '{"customer_id": 42}'

List all 32 built-in nodes:

m9m node list

30+ Integrations

What you need What's available
Store data PostgreSQL, MySQL, SQLite
Cloud storage AWS S3, GCP Cloud Storage, Azure Blob
Get AI help OpenAI (GPT-4), Anthropic Claude
Notify people Slack, Discord, Email (SMTP)
Manage code GitHub, GitLab
Spreadsheets Google Sheets
Read/write files Binary file I/O with encoding support
Any API HTTP Request node + Webhooks

Write custom logic in JavaScript or Python when you need it. See full node details with m9m node list.


Why m9m?

Zero infrastructure. One binary, no Node.js, no database required. Download it. Run it. You're done. Under 500ms from cold start to executing workflows.

Fast. Workflows run 5-10x faster than Node.js-based alternatives. Process 500 items in 6 seconds, not 60. Run m9m benchmark to see the numbers on your own hardware.

Free enterprise features. Git-based workflow versioning, audit logs, multi-workspace support, and Prometheus metrics are included -- not locked behind a paid tier.

Works with your existing n8n workflows. Import them and run them unchanged (see below).

m9m Typical Node.js alternative
Startup time 500ms 3s
Memory usage ~150MB ~512MB
Container size 300MB 1.2GB
Concurrent workflows 500 50

Works with n8n

Already using n8n? Import your workflows. They run unchanged, just faster.

m9m exec my-n8n-workflow.json

What's compatible:

  • Workflow JSON format and node connections
  • Expression syntax ({{ $json.field }}, {{ $node["name"].data }})
  • 32 node types covering the most-used n8n functionality
  • Trigger types (webhooks, cron schedules)

See the migration guide for details.


Go Further

These features are available when you need them. Start simple, add complexity later.

  • Run AI agents in sandboxed environments -- Claude Code, Codex, and Aider with resource limits and namespace isolation. Learn more
  • Embed in your app -- Go, Python, and Node.js SDKs for programmatic workflow execution. Learn more
  • Enterprise monitoring -- Prometheus metrics and OpenTelemetry tracing out of the box. Learn more
  • Build custom nodes -- Extend m9m with your own node types in Go. Learn more
  • MCP integration -- 37 tools for AI-powered workflow management via Claude Code. Learn more

Documentation

Resource Description
Getting Started Quick start guide
Architecture System design
API Reference REST API documentation
Node Development Building custom nodes
CLI Nodes CLI agent orchestration
Deployment Production deployment
MCP Integration Claude Code integration

Contributing

We welcome contributions! See our Contributing Guide for details.

git clone https://github.com/neul-labs/m9m.git
cd m9m
make deps && make test && make build

License

MIT License. See LICENSE for details.


Documentation | GitHub | Report an Issue

About

Agentic Workflow Automation at Ludicrous Speed

Topics

Resources

License

Contributing

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors