MCP Server

The Indigo MCP server exposes your meeting intelligence — signals, meetings, transcripts, and organizational taxonomy — through the Model Context Protocol. Connect it to Claude and query your data with natural language.

How it works

The MCP server acts as a bridge between your AI client and Indigo’s data layer. Queries are authenticated with your Indigo credentials and automatically scoped to your account — you only see your own data.

What’s exposed

4 collections

Signals, meetings, meeting data, and taxonomy — each queryable with filters, projections, sorting, and pagination.

4 tools

query_collection, aggregate_collection, list_collections, and collection_info for full data access.

Real-time data

Every query hits live data. No caching, no stale results. New signals appear immediately after meetings.

Multi-tenant isolation

Tenancy filters are injected automatically. Your queries can never access another user’s data.

Supported clients

Client	Transport	Status
Claude Desktop	StreamableHTTP	Supported
Claude Code	StreamableHTTP	Supported
Any MCP client	StreamableHTTP	Compatible

Setup

Run the one-command setup from the Indigo CLI:

indigo setup mcp

This authenticates with Indigo, generates the MCP server configuration, and writes it to your Claude Desktop config file. See the full setup guide for details.

Next steps

Data model

Understand the 4 collections and their schemas. Collections

Available tools

Reference for query and aggregation tools. Tools

Use cases

Practical prompts for common workflows. Use cases