Solace Agent Mesh

Open-source framework for building event driven multi-agent AI systems

Star ⭐️ this repo to stay updated as we ship new features and improvements.

Key Features • Quickstart • Next Steps

---

Whether you're prototyping an 🤖 AI assistant or deploying a 🌎 production-grade solution, Solace Agent Mesh (SAM) provides the infrastructure to:

• Connect AI agents to real-world data sources and systems.
• Add gateways to expose capabilities via REST, a browser-based UI, Slack, and many more.
• Scale from local development to distributed, enterprise deployments.

---

✨ Key Features

• ⚙️ Modular, Event-Driven Architecture – All components communicate via events through a central event mesh, enabling loose coupling and high scalability.
• 🤖 Composable Agents – Combine specialized AI agents to solve complex, multi-step workflows.
• 🌐 Flexible Interfaces – Interact with SAM via the REST API, browser UI, or Slack Integration.
• 🧠 Built-in Orchestration – Tasks are automatically broken down and delegated across agents by a built-in orchestrator.
• 🧩 Plugin-Extensible – Add your own agents, gateways, or services with minimal boilerplate.
• 🏢 Production-Ready – Backed by Solace’s enterprise-grade event broker for reliability and performance.
• 🔧 Services – File storage, memory, and embeddings, all extensible, and built-in.

---

🚀 Quick Start (5 minutes)

Set up Solace Agent Mesh in just a few steps.

⚙️ System Requirements

To run Solace Agent Mesh locally, you’ll need:

• Python 3.10.16+
• pip (comes with Python)
• OS: MacOS, Linux, or Windows (with WSL)
• LLM API key (any major provider or custom endpoint)

💻 Setup Steps

# 1. (Optional) Create and activate a Python virtual environment
python3 -m venv venv
source venv/bin/activate

# 2. Install the Solace Agent Mesh
pip install solace-agent-mesh

# 3. Initialize a new project
mkdir my-agent-mesh && cd my-agent-mesh
solace-agent-mesh init        # Follow the steps in the interactive init

# 4. Build and run the project
solace-agent-mesh run -b      # Shortcut for build + run

Once running:

• Open the Web UI at http://localhost:5001 to talk with a chat interface.

Use the REST API directly via curl

curl --location 'http://127.0.0.1:5050/api/v1/request' \
  --form 'prompt="What is the capital of France?"' \
  --form 'stream="false"'

---

➡️ Next Steps

Want to go further? Here are some hands-on tutorials to help you get started:

🔧 Integration	⏱️ Est. Time	📘 Tutorial
🌤️ Weather Agent Build an agent that gives Solace Agent Mesh the ability to access real-time weather information.	~5 min	Weather Agent Plugin
🗃️ SQL Database Integration Enable Solace Agent Mesh to answer company-specific questions using a sample coffee company database.	~10–15 min	SQL Database Tutorial
🧠 MCP Integration Integrating a Model Context Protocol (MCP) Server into Solace Agent Mesh.	~10–15 min	MCP Integration Tutorial
💬 Slack Integration Chat with Solace Agent Mesh directly from Slack.	~20–30 min	Slack Integration Tutorial

📚 Want to explore more? Check out the full Solace Agent Mesh documentation.

---

📦 Release Notes

Stay up to date with the latest changes, features, and fixes.
See CHANGELOG.md for a full history of updates.

---

👥 Contributors

Solace Agent Mesh is built with the help of our amazing community.
Thanks to everyone who has contributed ideas, code, and time to make this project better.
👀 View the full list of contributors → GitHub Contributors 🤝 Looking to contribute? Check out CONTRIBUTING.md to get started and see how you can help.

---

📄 License

This project is licensed under the Apache 2.0 License.
See the full license text in the LICENSE file.