​

MeshAgent

MeshAgent is your end-to-end platform for building, deploying, and operating AI agents — collaboratively, securely, and at any scale.

MeshAgent removes the infrastructure headaches of real-time agent collaboration by spinning up secure Rooms that connect humans, agents, tools, and shared context. Paired with a developer-friendly SDK, a browser-based Studio, and a fully-managed Cloud service (MeshAgent Server), MeshAgent lets you launch, share, and refine agents in hours instead of weeks.

​

Core building blocks

​

Why MeshAgent?

MeshAgent handles the complex infrastructure of running collaborative AI agents, so your teams can focus on building solutions that grow your business.

• AI Agents Integrated with Teams: MeshAgent’s unique Room architecture lets multiple people and agents share context and collaborate in real time. Secure Rooms spin up when the first participant arrives and vanish when the last one leaves, making collaboration seamless, continuous, and asynchronous — not a series of hand-offs.
• Accelerate Time to Value: Invite teammates to a secure, access-controlled Room in MeshAgent Studio where agents can be tested instantly in any browser. Engineers can update agent logic on-the-fly while stakeholders interact live, and front-end teams can simultaneously build the UI. This parallel workflow significantly shortens development cycles and tightens feedback loops. Once the agent is complete, easily deploy it and publish a shareable URL, giving authorized users direct access without requiring code checkouts, IDEs, or GPUs.
• Operate at any Scale with Enterprise Grade Governance: MeshAgent provides a uniform way to create, deploy, and manage agents. It automatically provisions and retires secure Rooms as participants come and go, scaling agents and collaboration spaces up or down as demand changes whether you’re running one agent or ten thousand.
• Reusable and Extensible Agent Tools: MeshAgent ships prebuilt, reusable toolkits for common Room interactions (e.g., file and document handling, media generation, dynamic UI interactions) and high-demand tasks (e.g., web search or document-to-Markdown conversion). Support for custom tools and the Model Context Protocol (MCP) enables you to seamlessly integrate any MCP-compatible service into a Room without changes to your applications. Tools are centrally managed like agents, allowing you to deploy them once and reuse them effortlessly.
• Enterprise Grade Observability: Streaming logs and OpenTelemetry traces capture every prompt, message, tool call, and error in real-time, enabling teams to proactively monitor, diagnose, and debug agent issues in minutes.
• Cost Management and Clarity: Clear usage and billing views break down minutes and credits by project, Room, and agent, enabling transparent cost tracking and accurate internal chargebacks or proactive budget alerts.
• Fits Your Tech Stack: Official SDKs for Python, JavaScript/TypeScript, and Dart expose the same MeshAgent feature set (Rooms, Agents, Tools, Messaging, Queues, Storage, Documents, etc.) so teams can stay productive in their preferred stack.

Because MeshAgent bundles collaboration, runtime, observability, scalability, and cost management into a single cohesive platform, you can move from idea to running, observable agent by the end of the day—not the end of the quarter.

​

Next Steps

1. Get set up with MeshAgent: Check out the Getting Started guide to install MeshAgent, connect to your first room, and run your first agent.

2. Watch the Getting Started with MeshAgent video series: These tutorials walk you through setting up MeshAgent, using the MeshAgent studio, building, deploying, and sharing your first agents. Check out the getting started series below or watch and subscribe on YouTube.

3. Explore the examples: Browse or clone our agent examples to kickstart your agent projects.

​

Getting Started Videos

​

Part 1: Setting up MeshAgent

In this video you will learn how to setup your MeshAgent account, install MeshAgent, generate your API keys, and call your first chat and voice agent into a secure MeshAgent room.

​

Part 2: Building Chat and Voice Agents with MeshAgent

In this video you will learn how to build a chat and voice agent in three phases. First you will build agents without tools, next you will add built-in MeshAgent tools to your agents, and finally you will add a custom tool to the agents. You will experiment with and test each of the agents using the MeshAgent studio.

​

Part 3: Deploy and Share Agents with MeshAgent

In this video you will learn how to deploy your agent as a MeshAgent service and generate a URL to share your agent with others.