Low code tool to rapidly build and coordinate multi-agent teams

Table of Contents

• What is Tribe?
• What are some use cases?
• Highlights
• How to get started
• Generate Secret Keys
• Deploy Tribe locally with Docker (simplest way)
• Deploy Tribe on a remote server
• Guides and concepts
• Contribution
• Release Notes
• License

What is Tribe?

Ever heard the saying, “Two minds are better than one?” This principle applies to agents as well! Tribe leverages the langgraph framework to allow you to customize and coordinate teams of agents effortlessly. By dividing complex tasks among specialized agents, each can focus on what it does best, resulting in faster and higher-quality solutions.

What are some use cases?

By teaming agents together, you can tackle greater challenges. Here are a few use cases:

• Football Analysis: One agent monitors the latest Premier League news while another analyzes the data and generates insightful reports.
• Trip Planning: One agent suggests the best restaurants, while another finds highly-rated hotels, covering every aspect of your vacation.
• Customer Service: Agents manage IT issues, complaints, and product inquiries, speeding up service with specialized expertise.

Highlights

• Persistent conversations for maintaining chat histories.
• Real-time observability to monitor agent performance.
• Tool calling that enables agent interaction with external APIs.
• Retrieval Augmented Generation (RAG) for reasoning with internal knowledge bases.
• Human-in-the-loop for approvals prior to tool calling.
• Support for open-source models.
• Easy deployment with Docker.
• Multi-tenancy for user and team management.

How to get started

Before diving into deployment, ensure you replace the following environment variable values in your setup:

• SECRET_KEY
• FIRST_SUPERUSER_PASSWORD
• POSTGRES_PASSWORD

Generate Secret Keys

Some environment variables have default values; use secure ones instead. To generate secret keys, run:

bash
python -c "import secrets; print(secrets.token_urlsafe(32))"

Repeat the above command to generate additional secure keys.

Deploy Tribe locally with Docker (simplest way)

Get up and started within minutes on your local machine.

Deploy Tribe on a remote server

Learn how to deploy Tribe on your remote server.

Guides and concepts

Sequential vs Hierarchical workflows

Sequential workflows

In a sequential workflow, agents complete tasks one after another based on dependencies. Think of it as a relay race where each runner’s performance impacts the next runner’s success.

Hierarchical workflows

In a hierarchical workflow, agents form a team-like structure led by a team leader who delegates smaller tasks to members, similar to a project manager overseeing a group with diverse expertise.

Skills

Skills are abilities assigned to agents for world interaction. Tribe offers skills for web searches, Wikipedia queries, and finance news retrieval, among others. You can also create custom skills:

Create a Skill Using Skill Definitions

You can define a skill by navigating to the Skills tab and providing structured skill definitions. For example:

{
  "url": "https://example.com",
  "method": "GET",
  "headers": {},
  "type": "function",
  "function": {
    "name": "Your skill name",
    "description": "Your skill description",
    "parameters": {
      "type": "object",
      "properties": {
        "param1": {
          "type": "integer",
          "description": "Description of the first parameter"
        },
        "param2": {
          "type": "string",
          "enum": ["option1"],
          "description": "Description of the second parameter"
        }
      },
      "required": ["param1", "param2"]
    }
  }
}

Writing a Custom Skill using LangChain

For more advanced tasks, use LangChain to create intricate tools and integrate them into Tribe. Remember to restart the application to load new custom skills properly.

Retrieval Augmented Generation (RAG)

RAG allows your agents to enhance their knowledge using additional data. Upload your private data for agents to reason upon for specialized tasks.

Troubleshooting

If you face difficulties during setup or operation, consider the following:

• Ensure all environment variables are correctly set.
• Check if Docker is installed and configured properly.
• Consult deployment logs for clues on any issues.

For more insights, updates, or to collaborate on AI development projects, stay connected with fxis.ai.

At fxis.ai, we believe that such advancements are crucial for the future of AI, as they enable more comprehensive and effective solutions. Our team is continually exploring new methodologies to push the envelope in artificial intelligence, ensuring that our clients benefit from the latest technological innovations.

Contribution

Tribe is open-sourced and welcomes contributions! Dive into our contribution guide to find out how you can help improve the project.

Release Notes

For details on changes and updates, check the release-notes.md file.

License

Tribe is licensed under the terms of the MIT license.