π Thank you to everyone who has built with Solace Agent Mesh! This Python version is now deprecated β it's no longer under active development and won't receive new features, bug fixes or security updates.
π Check out the new version of Solace Agent Mesh β https://docs.solace.com/Agent-Mesh/agent-mesh.htm
π₯οΈ A free edition of the Solace Agent Mesh desktop app is available: https://solace.com/products/agent-mesh/download
This repository will be archived (read-only) but stays available for reference, so your existing links and installs keep working.
Running Agent Mesh Enterprise from Wheel File
You can run Agent Mesh Enterprise directly from a Python wheel file without using Docker containers. This approach gives you direct control over your Python environment and integrates with existing Python-based deployments.
Prerequisitesβ
To run Agent Mesh Enterprise from a wheel file, you need:
- Python 3.10.16 β 3.13.x
- pip or uv package manager
- Access to the Solace Product Portal
- An LLM service API key and endpoint (optionalβyou can configure models through the Agent Mesh UI after starting the application)
- For production deployments, Solace broker credentials
Step 1: Download the Wheel Fileβ
Download the Agent Mesh Enterprise wheel file from the Solace Product Portal.
The wheel file follows the naming pattern:
solace_agent_mesh_enterprise-<version>-py3-none-any.whl
Step 2: Install the Wheel Fileβ
You can install the wheel file with either pip or uv.
If you use pip, run:
pip install solace_agent_mesh_enterprise-<version>-py3-none-any.whl
If you use uv, run:
uv pip install solace_agent_mesh_enterprise-<version>-py3-none-any.whl
This installation provides the solace-agent-mesh CLI tool and the Agent Mesh Enterprise framework.
Step 3: Prepare Your Configurationβ
Agent Mesh Enterprise requires configuration files that define your agents, gateways, and system settings. The solace-agent-mesh init command creates the project directory structure, environment variables, and configuration files you need.
Follow the project setup steps in the Creating and Running an Agent Mesh Project guide to initialize your configuration. You can use the web-based interface or command-line options to configure your project settings.
After initialization, your project directory contains the necessary configs/ directory with agent and gateway configurations, plus a .env file with your credentials.
Step 4: Run Agent Mesh Enterpriseβ
Start Agent Mesh Enterprise using the solace-agent-mesh run command:
solace-agent-mesh run
This command loads environment variables from your .env file, starts all agents and gateways defined in your configs/ directory, and launches the web UI if you configured one.
Running Specific Componentsβ
To run only specific agents or gateways, provide the configuration files as arguments:
solace-agent-mesh run configs/agents/orchestrator.yaml configs/gateways/webui.yaml
Using the Short Aliasβ
The sam alias provides a shorter alternative to the full command name:
sam run
Limitationsβ
When running from a wheel file, certain features have limitations compared to the Docker-based deployment.
No Dynamic Agent Deploymentβ
Dynamic agent deployment through the UI or API is not supported when running from a wheel file. You cannot deploy new agents at runtime through the web interface, upload agent configurations through the API, or dynamically add or remove agents without restarting. All agents must be defined in configuration files before you start the application.
Custom Agent Codeβ
To create custom agents with Python tools and lifecycle functions, follow the standard agent creation process. For detailed instructions on creating agents with custom code, see Creating Agents.
Accessing the Web UIβ
After starting Agent Mesh Enterprise, access the web interface at the configured port (default: http://localhost:8000).
If you specified a different port in your gateway configuration, use that port instead.
Next Stepsβ
After running Agent Mesh Enterprise from the wheel file, you can configure authentication and authorization for your deployment. For information about setting up authentication, see Single Sign-On. For information about configuring authorization, see Role-Based Access Control. For production deployment considerations, see deployment options.