The Agent Management Studio (AMS) is the visual, no-code admin portal for building, testing, and monitoring stateful AI agents through our SDK or API. The AMS enables you to easily control every aspect of your agent’s operation, from tool execution, MCP server management, data access, guardrails, personality, and more. Built from the ground up with a focus on simplicity and efficiency, AMS is laser-focused on what matters most: creating, managing, and deploying stateful agents.

​
Why use the AMS?

  • Collaboration: You and your team can build & improve agents in a shared space.
  • Simple management: No need to code - build complex workflows easily, add new tools, change agent behavior, and more.
  • Easy debugging and improving: Live dashboards let you spot misbehaving agents or workflows quickly and make fast improvements.
  • Rapid prototyping: Create and test agents in a fraction of the time, then put them in front of real users for quick feedback.
AMS features the new workflows that matter: agent management, context handling, API management, and our breakthrough workflow builder that lets you create custom agent functionality by simply talking to an agent.

​
AMS Key Components

​
Agents

Agents are intelligent, autonomous units designed to perform specific tasks, make decisions based on their defined role and objectives, and use available workflows to achieve outcomes. Each agent acts like a subject matter expert, tailored for a particular function. Learn more about Agents

​
Knowledge Contexts

Knowledge Context is a foundational system that equips agents with access to the right information when they need it. It works like a dynamic reference library, bundling relevant content together so agents can retrieve facts, files, or insights while helping users. Learn more about Knowledge Contexts

​
MCP Servers

MCP stands for Model Context Protocol - think of it as a “universal translator” that lets AI agents talk to different tools, data sources, and systems in a consistent way. MCP servers standardize how AI agents exchange information with external systems.
It helps you bundle information, workflows, and tools (actions and information agents can access) into one system that your agents can use. Learn more about MCP Servers

​
Workflows

Workflows transform how you build agent capabilities - instead of complex development, you describe what you want your agent to do in natural conversation, and the system generates functional tools automatically. Learn more about Workflows

​
Dashboard

The Dashboard is where you can preview and deploy your agents.
You can also monitor and interact with live agents, manage your human collaborators, and handle your APIs. Learn more about Deployment