What is Mule?

Mule is an advanced multi-agent AI development platform that orchestrates sophisticated workflows and integrations. It serves as your autonomous development infrastructure, capable of:

  • Multi-Agent Workflows: Orchestrate complex tasks with sequential steps, sub-workflows, and validation
  • Production Integrations: Discord, Matrix, gRPC API, RSS feeds, persistent memory, and more
  • Multi-Provider Support: OpenAI, Anthropic, Gemini, and Ollama with dynamic model switching
  • Autonomous Development: Monitor repositories, process issues, create PRs, and respond to feedback
  • Advanced Memory: ChromeM-based vector database for RAG and context retention
  • Quality Control: Built-in validation functions with retry logic for robust outputs

Prerequisites

Before installing Mule, ensure you have:

  • Go 1.24+: Mule is built with Go 1.24 or newer
  • Git: For repository operations
  • SSH keys: Configured for GitHub access if using GitHub repositories
  • AI provider access: API keys for your chosen AI provider (OpenAI, Anthropic, Gemini, Ollama)
  • GitHub token: If you plan to use GitHub repositories (optional)

Installation

Standard Installation 

git clone https://github.com/mule-ai/mule.git
cd mule
make run

Docker Installation 

git clone https://github.com/mule-ai/mule.git
cd mule
make docker-build
make docker-run

Installing Dependencies

For development or contributing to Mule:

make download-golangci-lint  # Install linting tools
make deps                    # Install Go dependencies

Configuration

Mule stores its configuration in a standard location based on your operating system:

  • Linux: ~/.config/mule/
  • macOS: ~/Library/Application Support/mule/
  • Windows: %APPDATA%\mule\

The primary configuration is automatically created on first run.

Running Mule

After installation, Mule is accessible through a web interface at http://localhost:8083.

To get started running Mule, follow these steps:

  1. Set a GitHub token (optional)

    • This allows Mule to interact with GitHub repositories
    • Required for working with private repositories

  2. Set up your AI Providers

    • Configure at least one AI provider
    • Supports OpenAI, Anthropic, Gemini, and Ollama
    • Provide necessary API keys and endpoints

  3. Create Agents

    • Configure agents with appropriate models and prompts
    • Set up the tools agents can access
    • Define validation functions to ensure code quality

  4. Configure the System Agent

    • Handles system tasks like commit messages and PR descriptions
    • Uses the same AI providers as regular agents

  5. Add a Repository

    • Connect to GitHub repositories or use local ones
    • Set up sync schedules to monitor for issues

  6. Create an Issue

    • In GitHub: Create an issue and add the mule label
    • With Local Provider: Use the Mule interface to create an issue

Once set up, Mule will automatically:

  1. Monitor repositories on the schedule you defined
  2. Find issues labeled with mule
  3. Generate solutions and create pull requests
  4. Respond to any comments on those pull requests

Command Line Options

Mule supports several command line options:

mule [options]

Options:

  • --config: Specify a custom config path
  • --port: Set a custom port (default: 8083)
  • --debug: Enable debug logging
  • --help: Display help information

Demo

See Mule in action with this demonstration using the Local Provider:

Next Steps

After getting Mule running, you might want to:

Troubleshooting

Common Issues

Mule can’t connect to GitHub

  • Verify your GitHub token has the required permissions
  • Check network connectivity and firewall settings

AI provider connection fails

  • Ensure API keys are correctly entered
  • Verify endpoint URLs are accessible
  • Check rate limits on your AI provider

Repository sync fails

  • Ensure SSH keys are properly configured
  • Verify repository path permissions
  • Check for available disk space

No changes generated

  • Review issue description for clarity
  • Check agent prompt templates
  • Verify that selected AI model is capable of code generation

For more help, check the logs at ~/.config/mule/mule.log or use the Logs page in the web interface.