Lets AI clients build, configure, and run multi-agent AI assistants and workflows through a visual builder or TypeScript SDK.
Also known as inkeep/agents
- Stars
- 1,499
- Tools
- —
- Downloads, 30 days
- —
- Latest version
- —
- Last commit
- Nov 13, 2026
- License
- —
What is Inkeep Agents MCP?
Inkeep Agents lets AI clients build, configure, and run multi-agent AI assistants and workflows through a visual builder or TypeScript SDK.
Summary generated from the project's README
About Inkeep Agents MCP server
Building and managing multi-agent AI assistants is possible through this platform using either a no-code visual builder or a TypeScript SDK. The system allows technical and non-technical team members to collaborate because the visual drag-and-drop canvas and the code-first framework maintain full two-way sync. Developers can use the @inkeep/agents-sdk to define agents and sub-agents with type safety and intellisense, while the agents-cli provides push and pull utilities to keep code and visual configurations aligned.
Users can deploy these agents using Docker or Vercel and interact with them via MCP, A2A, or Vercel SDK APIs. The architecture includes an API for managing credentials, projects, and MCP servers, alongside a UI component library for embedding chat interfaces into web applications. Observability is provided through a Traces UI and OpenTelemetry to monitor agent execution and conversation states.
What you can do with it
- 1Deploying customer experience assistantsYou can deploy agents to act as real-time chat assistants for help centers or technical documentation.
- 2Automating repetitive business workflowsDevelopers can use agents to handle tasks like updating CRMs or triaging helpdesk tickets.
- 3Managing internal company copilotsOrganizations can create specialized agents to assist support, sales, or marketing departments.
Written from the project's README. Every use case is checked against a quote from it.
Automated checks
Ran Nov 14, 2026Full report →Not verified yet. A server needs a score of 80, no serious finding, and enough data for us to check. How checks work · What to fix for more points
- Source repository linked and reachableinkeep/agents is public and reachablegithub.com/inkeep/agents ↗Pass
- Package names this repository as its sourceNot checked
- Build provenance (Sigstore attestation)Not checked
- Open source license (OSI)No licenseFinding
- Commit or release in the last yearLast commit or release 2026-11-11github.com/inkeep/agents ↗Pass
- OpenSSF ScorecardNot checked
- Package name not close to a popular packageNot checked
- No critical advisory in the package or its dependenciesNot checked
- No hidden instructions to the model0 tool descriptions and the README read; no override phrases, credential paths or hidden charactersPass
- Destructive tools are labelledNot checked
- Remote endpoint answers the MCP handshakeNot checked
README
View on GitHubInkeep Agents
Build AI Agents with a No-Code Visual Builder or TypeScript SDK. Agents can be edited in either with full 2-way sync, so technical and non-technical teams can create and manage Agents in one platform.
Get started with the docs or 1-minute quick start.
Two ways to build
No-Code Visual Builder
A drag-and-drop canvas so any team can create and own the Agents they care about.
TypeScript Agents SDK
A code-first framework so engineering teams can build with typesafety, intellisense, CI/CD, and the tools they expect.
import { agent, subAgent } from "@inkeep/agents-sdk";
import { consoleMcp } from "./mcp";
const helloAgent = subAgent({
id: "hello-agent",
name: "Hello Agent",
description: "Says hello",
canUse: () => [consoleMcp],
prompt: `Reply to the user and console log "hello world" with fun variations like h3llo world`,
});
export const basicAgent = agent({
id: "basic-agent",
name: "Basic Agent",
description: "A basic agent",
defaultSubAgent: helloAgent,
subAgents: () => [helloAgent],
});
The Visual Builder and TypeScript SDK are fully interoperable: technical and non-technical teams can edit and manage Agents in either format and collaborate with others at any time.
Use Cases
Inkeep Agents can operate as real-time AI Chat Assistants, for example:
- a customer experience agent for help centers, technical docs, or in-app experiences
- an internal copilot to assist your support, sales, marketing, ops, and other teams
Agents can also be used for AI Workflow Automation like:
- Creating and updating knowledge bases, documentation, and blogs
- Updating CRMs, triaging helpdesk tickets, and tackling repetitive tasks
Platform Overview
Inkeep Open Source includes:
- A Visual Builder & TypeScript SDK with 2-way sync
- Multi-agent architecture to support teams of agents
- MCP Tools with credential management
- A UI component library for dynamic chat experiences
- Triggering Agents via MCP, A2A, & Vercel SDK APIs
- Observability via a Traces UI & OpenTelemetry
- Easy deployment using Vercel or Docker
For a full overview, see the Concepts guide.
Interested in a managed platform? Sign up for the Inkeep Cloud waitlist or learn about Inkeep Enterprise.
Architecture
The Inkeep Agent Platform is composed of several key services and libraries that work together:
- agents-api: An API that handles configuration of Agents, Sub Agents, MCP Servers, Credentials, and Projects with a REST API. Additionally, it exposes Agent execution and evaluation. The API tracks conversation state and emits OTEL traces.
- agents-manage-ui: Visual Builder web interface for creating and managing Agents. Writes to the
agents-api. - agents-sdk: TypeScript SDK (
@inkeep/agents-sdk) for declaratively defining Agents and custom tools in code. Writes toagents-api. - agents-cli: Includes various handy utilities, including
inkeep pushandinkeep pullwhich sync your TypeScript SDK code with the Visual Builder. - agents-ui: A UI component library of chat interfaces for embedding rich, dynamic conversational AI experiences in web apps.
Under the hood, the framework uses the Vercel AI SDK for interfacing with LLM providers, so it's compatible with Vercel's useChat hook and other AI primatives.
License and Community
The Inkeep Agent Framework is licensed under the Elastic License 2.0 (ELv2) subject to Inkeep's Supplemental Terms (SUPPLEMENTAL_TERMS.md). This is a fair-code, source-available license that allows broad usage while protecting against certain competitive uses.
Inkeep is designed to be extensible and open: use the LLM provider of your choice, use Agents via standard protocols, and easily deploy and self-host Agents in your own infra.
If you'd like to contribute, follow our contribution guide.
Join our community to get support, stay up to date, and share feedback.
Inkeep Agents MCP questions
Can I build agents without writing code?
Yes. The No-Code Visual Builder provides a drag-and-drop canvas so non-technical teams can create agents.
Does the visual builder sync with the TypeScript SDK?
Yes, the Visual Builder and TypeScript SDK are fully interoperable with 2-way sync.
What can Inkeep Agents be used for?
They work as real-time AI chat assistants and for AI workflow automation tasks.
Can agents call external tools via MCP?
Yes, the platform supports MCP tools with credential management and triggering agents via MCP, A2A, and Vercel SDK APIs.
How are agents deployed?
You can deploy agents using Vercel or Docker and self-host them in your own infrastructure.