Indx Documentation

Indx is a high-performance search engine that uses pattern recognition instead of keyword matching. It handles typos, formatting variations, and messy input without configuration.

Two Ways to Run Indx

Embed it — the IndxSearchLib NuGet package. Search runs inside your .NET application: no server, no network hop, nothing else to deploy. For a single .NET app that owns its data.

Host it — the Indx server. A self-contained server with an HTTP API, a dashboard for teams, datasets and field configuration, API keys, and an MCP endpoint for AI agents. For any language, for several applications sharing one index, or when someone other than the developer manages the data. Self-host it, or deploy it from the Azure Marketplace.

AI Agent Skill

Use Indx with AI coding agents via the Indx Search skill on skills.sh. The skill covers both the embedded library and the server, field configuration, querying, filters, boosts, coverage tuning, and search UX patterns.

Get the AI skill

For the full agent-driven setup — skill, agent-readable docs, and connecting your agent to a live server over MCP — see Build with an AI Agent.

Guides

How To

Practical examples with both C# and HTTP API code:

API Reference

React UI Library

indx-react is a complete search UI library for React applications connecting to Indx. It covers the full search experience out of the box:

PackageDescription
@indxsearch/intrface
Ready-to-use search components: search input, results list, value filters, range filters, active filter panel, sort controls, and loading skeletons
@indxsearch/systm
Design system with buttons, inputs, sliders, checkboxes, and more — customisable via CSS variables
@indxsearch/indx-types
Full TypeScript types for the Indx v2 HTTP API — queries, filters, boosts, results, field configuration, and system status

Resources