Inkeep Platform — Product Suite Overview

Inkeep is an AI-powered support platform that helps teams build intelligent search and chat experiences on top of their own documentation. This article gives developers and technical teams a quick map of every product and integration layer available in the Inkeep platform.

Platform capabilities

The core Inkeep platform includes the following built-in capabilities:

• No-Code Studio — configure and deploy AI experiences without writing code.
• Multi-agent runtime — orchestrate multiple specialized AI agents to handle different types of questions.
• Retrieval-Augmented Generation (RAG) — ground AI answers in your own content so responses stay accurate and on-topic.
• Analytics — track what users ask, how the AI responds, and where gaps in your content exist.
• Enterprise security — role-based access controls, data isolation, and compliance-ready infrastructure.

Developer tools

TypeScript Agents SDK

The @inkeep/agents-sdk package lets you build and extend AI agents in TypeScript. Use it to integrate Inkeep's agent runtime into your own backend services or custom workflows.

CLI

The @inkeep/agents-cli command-line tool lets you manage agents, run local tests, and automate deployment tasks from your terminal.

UI Kit

The @inkeep/uikit package provides ready-made React components and plain JavaScript snippets for embedding AI chat and search directly into your product or documentation site. Drop in a chat widget, a search bar, or a full AI assistant — no backend setup required on your end.

APIs

Inkeep exposes three API surfaces for programmatic access:

AI Chat API

Send messages and receive AI-generated answers backed by your content. Use this to power chat interfaces, Slack bots, or any conversational surface.

Analytics API

Query usage data, question logs, and AI performance metrics. Use it to build custom dashboards or feed support insights into your existing data pipeline.

RAG API

The RAG API exposes Inkeep's retrieval-augmented generation model (inkeep-rag) through an OpenAI-compatible interface. If you already use the OpenAI SDK or any OpenAI-compatible client, you can switch in the inkeep-rag model with minimal code changes.

 

The inkeep-rag model follows the OpenAI Chat Completions API format, so you can use it as a drop-in replacement in tools that already support OpenAI-compatible endpoints.

MCP Server

Inkeep provides a Model Context Protocol (MCP) server so you can connect Inkeep's knowledge retrieval directly to MCP-compatible AI clients — including Cursor, Claude Desktop, and other MCP clients. Once connected, those tools can query your Inkeep knowledge base as a context source when generating code suggestions or answers.

To get started, point your MCP client at the Inkeep MCP server endpoint and authenticate with your API key. No additional SDK setup is needed for basic MCP usage.

Choosing the right integration

Not sure where to start? Here's a quick guide:

• Embedding chat or search in a web app → use the UI Kit (@inkeep/uikit).
• Building a custom backend or agent workflow → use the Agents SDK (@inkeep/agents-sdk).
• Calling the AI from an existing OpenAI-compatible client → use the RAG API with model inkeep-rag.
• Getting answers in Cursor, Claude Desktop, or another AI coding tool → use the MCP Server integration.
• Managing or testing agents from the command line → use the CLI (@inkeep/agents-cli).
• Pulling usage data into your own tools → use the Analytics API.