Skip to content

About

Platform for building and running AI agents across the full lifecycle — build, test, deploy, monitor, improve. Hosts your agents, runs their tools, and coordinates them over an event mesh, with sessions, artifacts, auth, evaluators, and observability built in.

Topics

Resources

Contributing

Security policy

Stars

4.9k stars

Watchers

22 watching

Forks

Latest commit

 

History

2,293 Commits

Folders and files

Repository files navigation

Solace Agent Mesh

Solace Agent Mesh is an agent development and runtime platform. It supports the full agent development lifecycle: build, test, deploy, monitor, and improve your agents and business process workflows. It hosts the agents you build, runs the tools they call, and coordinates everything over an event broker, with sessions, artifacts, authentication, evaluators, and observability built in.

Note

The Solace Agent Mesh source code is coming to this repository soon. Until then, this repository holds Agent Mesh Desktop downloads and ready-to-run examples.

Installing Agent Mesh Desktop

Agent Mesh Desktop is free, runs as a single application on your computer, and includes the sam command line interface (CLI).

To install the application, run the install command for your operating system. The command downloads the application from the latest release.

Agent Mesh Desktop runs on the following platforms:

  • macOS 11 (Big Sur) and later
  • Windows 10 and later (x86_64)
  • Linux (Experimental)

On macOS and Linux:

curl -fsSL https://raw.githubusercontent.com/SolaceProducts/solace-agent-mesh/main/install/install.sh | sh

On Windows, in PowerShell:

irm https://raw.githubusercontent.com/SolaceProducts/solace-agent-mesh/main/install/install.ps1 | iex

The script selects the download that matches your operating system and processor. On Linux, it also selects your package format and WebKitGTK version. Before it installs anything, the script checks the download against the SHA-256 checksum of the release asset, and verifies the Apple notarization on macOS or the Authenticode signature on Windows.

To preview what the script does without installing, run it with --dry-run. On macOS and Linux, pass the flag through sh -s --:

curl -fsSL https://raw.githubusercontent.com/SolaceProducts/solace-agent-mesh/main/install/install.sh | sh -s -- --dry-run

On Windows, set $env:SAM_INSTALL_DRY_RUN=1 instead.

You can also download the asset for your platform from the latest release and install it yourself.

To connect an LLM the first time you open the application, see Installing the Desktop Bundle.

Using the sam CLI

To run sam from a terminal, add it to your PATH on Windows, or create an alias on macOS and Linux:

  • On Windows: The installer places sam.exe in %LOCALAPPDATA%\Programs\Solace Agent Mesh\cli, and the install script adds that folder to your user PATH. Open a new terminal to use sam. If you install without the script, add the folder yourself.
  • On macOS: Add an alias: alias sam='"/Applications/Solace Agent Mesh.app/Contents/MacOS/sam-desktop"'
  • On Linux: Add an alias: alias sam='solace-agent-mesh'

The following examples show common sam commands. The sam ai-assistance and sam config commands are Early Access, so their commands, flags, and output may change between releases.

# Install skills that teach your AI coding assistant, such as Claude Code or
# Codex CLI, to build agents, tools, and configuration for your Agent Mesh
# version. By default, the skills are written to .agents/skills/ in the current
# directory, where Codex CLI reads them. For Claude Code, add --target claude.
sam ai-assistance skill install

# Preview a configuration change, and then apply it to a running Agent Mesh
# instance. Replace manifest.yaml with the path to your manifest.
sam config plan -m manifest.yaml
sam config apply -m manifest.yaml

# Open the Agent Mesh documentation offline
sam docs

For every command and flag, see the CLI reference. To list the top-level commands, run sam --help. To see the flags of a command, run sam <command> --help.

Examples

The examples directory contains use cases that you apply to Agent Mesh with sam config apply. In the support ticket triage use case, a customer creates a ticket in the fictional Acme Help Desk, and Agent Mesh classifies the issue, calculates the response deadline, drafts the first reply, escalates the ticket if it is urgent, and writes the result back to the ticket.

The examples are for learning and evaluation, and are not production-ready.

Moving to Production

Production use of Agent Mesh requires a Solace commercial license. Agent Mesh Desktop is for building and testing on your own computer. For production, deploy Agent Mesh to Kubernetes, or use Agent Mesh Cloud, a managed cloud service. For more information, see Installing Kubernetes for Production and Agent Mesh Cloud. To discuss a production deployment with a Solace expert, book a demo.

License

For the license that covers each part of this repository, see NOTICE.

Getting Help

This repository does not accept issues or pull requests. For more information, see CONTRIBUTING.md.

About

Platform for building and running AI agents across the full lifecycle — build, test, deploy, monitor, improve. Hosts your agents, runs their tools, and coordinates them over an event mesh, with sessions, artifacts, auth, evaluators, and observability built in.

Topics

Resources

Contributing

Security policy

Stars

4.9k stars

Watchers

22 watching

Forks

Releases

Used by

Contributors

Languages