A monorepo package for MCP tools with dynamic loading capabilities - featuring Evolution API v2 WhatsApp integration
Drop any OpenAPI spec → Get an intelligent MCP agent that learns how you work and gets better with every interaction. Export to code when you're ready to customize.
Born from our daily work at Namastex Labs, AutoMagik Tools creates self-evolving agents that turn any API into a natural language interface.
Unlike static tools, AutoMagik agents remember and adapt:
# First time: "How much did I sell last month?"
# Agent learns your sales endpoints, date formats, and preferences
# By the 10th interaction:
# Agent already knows exactly which data you want and how you like it formatted
uvx automagik-tools tool automagik -t sse --port 8000
Three Intelligence Modes:
Powered by GPT-4.1 family models for cost-effective agentic behavior and reliable API reasoning.
Turn any OpenAPI spec into an intelligent agent:
# Discord API becomes a smart agent
uvx automagik-tools openapi \
https://raw.githubusercontent.com/discord/discord-api-spec/main/specs/openapi.json \
-t sse --port 8001
# Share this agent with your entire team via SSE
# Team members can access the same learning agent at http://localhost:8001
Now you can say:
Agent learns your patterns, server preferences, and communication style.
Genie connects any MCP servers and orchestrates them with persistent memory:
# Run Genie with memory-based agents
uvx automagik-tools tool genie -t sse --port 8000
Genie can orchestrate any combination of MCP servers. Configure via environment variables:
# Method 1: JSON configuration for multiple servers
export GENIE_MCP_CONFIGS='{
"agent-memory": {
"url": "http://192.168.112.149:8000/sse",
"transport": "sse"
},
"filesystem": {
"command": "npx",
"args": ["-y", "@modelcontextprotocol/server-filesystem", "/path/to/allowed/directory"],
"env": {}
},
"github": {
"command": "uvx",
"args": ["mcp-server-git"],
"env": {"GITHUB_TOKEN": "your-token"}
}
}'
# Method 2: AutoMagik-specific shorthand
export GENIE_AUTOMAGIK_API_KEY="your-api-key"
export GENIE_AUTOMAGIK_BASE_URL="http://localhost:8881"
export GENIE_AUTOMAGIK_TIMEOUT="600"
{
"mcpServers": {
"genie": {
"command": "uvx",
"args": [
"automagik-tools@latest",
"serve",
"--tool",
"genie",
"--transport",
"stdio"
],
"env": {
"OPENAI_API_KEY": "your-openai-key",
"GENIE_MCP_CONFIGS": "{\"agent-memory\":{\"url\":\"http://192.168.112.149:8000/sse\",\"transport\":\"sse\"},\"filesystem\":{\"command\":\"npx\",\"args\":[\"-y\",\"@modelcontextprotocol/server-filesystem\",\"/allowed/path\"],\"env\":{}}}"
}
}
}
}
Now in Claude/Cursor, Genie can:
Enterprise-grade agent orchestration that speaks human:
uvx automagik-tools tool automagik --transport sse --port 8000
Real examples from our users:
💬 Natural Language • 🧠 Memory & Learning • 🔄 Task Orchestration • 🏗️ Framework Agnostic • 👩💻 Export to Code
Create a .env file with your API keys:
# .env
OPENAI_API_KEY=sk-your-openai-key-here
AUTOMAGIK_API_KEY=your-automagik-api-key
AUTOMAGIK_BASE_URL=http://localhost:8881
# Enable JSON to Markdown processing (optional)
ENABLE_JSON_PROCESSING=true
JSON_PROCESSOR_MODEL=gpt-4.1-nano
{
"mcpServers": {
"automagik": {
"command": "uvx",
"args": [
"automagik-tools@latest",
"serve",
"--tool",
"automagik",
"--transport",
"stdio"
],
"env": {
"AUTOMAGIK_API_KEY": "YOUR_API_KEY",
"AUTOMAGIK_BASE_URL": "http://localhost:8881",
"OPENAI_API_KEY": "YOUR_OPENAI_API_KEY"
}
}
}
}
Where to add:
~/.cursor/mcp.json# Jira becomes conversational project management
OPENAI_API_KEY=your_key JIRA_API_TOKEN=your_token uvx automagik-tools serve \
--openapi-url https://dac-static.atlassian.com/cloud/jira/platform/swagger-v3.v3.json \
--transport sse --port 8002
# Shopify for e-commerce automation
OPENAI_API_KEY=your_key SHOPIFY_ACCESS_TOKEN=your_token uvx automagik-tools serve \
--openapi-url https://shopify.dev/docs/api/admin-rest/2023-04/openapi.json \
--transport sse --port 8003
# SSE mode allows your team to share the same learning agent
Personal Automation:
Team Coordination:
Business Intelligence:
The agent learns your patterns - after a few interactions, it knows exactly how you like your data formatted, which metrics matter most, and when to proactively alert you.
Configure Genie with different MCP server combinations:
# Development setup with local tools
export GENIE_MCP_CONFIGS='{
"filesystem": {
"command": "npx",
"args": ["-y", "@modelcontextprotocol/server-filesystem", "/home/user/projects"],
"env": {}
},
"git": {
"command": "uvx",
"args": ["mcp-server-git"],
"env": {"GIT_AUTHOR_NAME": "Your Name"}
}
}'
# Production setup with external services
export GENIE_MCP_CONFIGS='{
"automagik": {
"command": "uvx",
"args": ["automagik-tools", "tool", "automagik", "--transport", "stdio"],
"env": {
"AUTOMAGIK_API_KEY": "prod-key",
"AUTOMAGIK_BASE_URL": "https://api.yourcompany.com"
}
},
"slack": {
"command": "docker",
"args": ["run", "-i", "--rm", "-e", "SLACK_BOT_TOKEN", "mcp/slack"],
"env": {"SLACK_BOT_TOKEN": "xoxb-your-token"}
}
}'
You can also pass MCP server configurations directly to Genie via the ask_genie tool:
{
"tool": "ask_genie",
"arguments": {
"query": "List all my GitHub repositories and remember my coding preferences",
"mcp_servers": {
"github": {
"command": "uvx",
"args": ["mcp-server-git"],
"env": {"GITHUB_TOKEN": "your-token"}
},
"memory": {
"url": "http://localhost:8000/sse",
"transport": "sse"
}
}
}
}
The old way: Hours writing API integrations, maintaining complex schemas, fighting with documentation. When APIs change, everything breaks.
The AutoMagik way:
When your needs evolve, your agent learns and remembers.
AI orchestration that speaks human:
# Quick test with SSE
uvx automagik-tools tool automagik --transport sse --port 8000
What you can do:
Universal MCP orchestrator with persistent memory:
# Run as SSE server for team sharing
uvx automagik-tools tool genie --transport sse --port 8000
Capabilities:
Smart Claude workflow orchestration with real-time progress tracking:
# Execute Claude Code workflows with progress monitoring
uvx automagik-tools tool automagik-workflows --transport stdio
Features:
Complete WhatsApp automation:
Our roadmap includes agents that:
Production Deployments (Coming Soon):
Coming 2025: A hive of AI agents maintaining this entire codebase - reporting bugs, implementing features, and keeping users happy. All open source.
{
"mcpServers": {
"genie_with_memory": {
"command": "uvx",
"args": [
"automagik-tools@latest",
"serve",
"--tool",
"genie",
"--transport",
"stdio"
],
"env": {
"OPENAI_API_KEY": "your-openai-key",
"GENIE_MCP_CONFIGS": "{\"agent-memory\":{\"url\":\"http://192.168.112.149:8000/sse\",\"transport\":\"sse\"}}"
}
}
}
}
# Stripe Payments
uvx automagik-tools serve \
--openapi-url https://raw.githubusercontent.com/stripe/openapi/master/openapi/spec3.json \
--api-key $STRIPE_API_KEY
# GitHub API
uvx automagik-tools serve \
--openapi-url https://raw.githubusercontent.com/github/rest-api-description/main/descriptions/api.github.com/api.github.com.json \
--api-key $GITHUB_TOKEN
# Your Internal API
uvx automagik-tools serve \
--openapi-url https://api.yourcompany.com/openapi.json \
--api-key $YOUR_API_KEY
# Clone the repo
git clone https://github.com/namastexlabs/automagik-tools
cd automagik-tools
# Install with all dev dependencies
make install
# Run tests
make test
# Create a new tool
make new-tool
# Method 1: Dynamic (no files created)
uvx automagik-tools openapi https://api.example.com/openapi.json
# Method 2: Generate persistent tool
uvx automagik-tools create-tool --url https://api.example.com/openapi.json --name my-api
uvx automagik-tools tool my-api
automagik_tools/tools/your_tool/__init__.py with FastMCP serverSee our Tool Creation Guide for details.
# Core commands
automagik-tools list # List all available tools
automagik-tools hub # Serve all tools together
automagik-tools tool <name> # Serve a specific tool
automagik-tools openapi <url> # Serve from OpenAPI spec
automagik-tools mcp-config <tool> # Generate MCP config
automagik-tools info <tool> # Show tool details
automagik-tools version # Show version
# Development commands
make install # Install dev environment
make test # Run all tests
make lint # Check code style
make format # Auto-format code
make build # Build package
make docker-build # Build Docker images
We love contributions! See CONTRIBUTING.md for guidelines.
MIT License - see LICENSE
Built with ❤️ by Namastex Labs
Special thanks to:
Every API becomes a smart agent that learns how you work.
Star us on GitHub •
Join our Discord •
Follow on Twitter
FAQs
A monorepo package for MCP tools with dynamic loading capabilities - featuring Evolution API v2 WhatsApp integration
We found that automagik-tools demonstrated a healthy version release cadence and project activity because the last version was released less than a year ago. It has 1 open source maintainer collaborating on the project.
Did you know?
Socket for GitHub automatically highlights issues in each pull request and monitors the health of all your open source dependencies. Discover the contents of your packages and block harmful activity before you install or update your dependencies.
Security News
npm now supports Trusted Publishing with OIDC, enabling secure package publishing directly from CI/CD workflows without relying on long-lived tokens.
Research
/Security News
A RubyGems malware campaign used 60 malicious packages posing as automation tools to steal credentials from social media and marketing tool users.
Security News
The CNA Scorecard ranks CVE issuers by data completeness, revealing major gaps in patch info and software identifiers across thousands of vulnerabilities.