Entity Enricher ships a hosted Model Context Protocol (MCP) server, so AI agents enrich data by talking to it -- no scripts, no CSV round-trips. Connect it to claude.ai, Claude Desktop, Claude Code, or Cursor and your assistant can author schemas, run single or batch enrichments, merge model results, and read the records back, all inside one conversation.
The Model Context Protocol is the open standard that lets AI assistants call external tools. An MCP data enrichment server exposes enrichment operations as tools an agent can invoke: instead of you exporting a spreadsheet, running a pipeline, and importing results, you tell your assistant "enrich these 50 companies against my Company schema" and it orchestrates the work -- starting jobs, polling progress, and answering follow-up questions along the way.
Entity Enricher's MCP server is embedded in the platform itself -- the same engine behind the web app and REST API, with the same role-based permissions. It exposes 29 tools and 2 resources covering schema authoring, sample generation, single and batch enrichment, multi-model fusion, model benchmarks, and async job control.
Generate a JSON schema from a sample or a description, refine it in natural language, and save it -- the same self-correcting generator as the web app.
Start one enrichment or a parallel batch, poll job status, cancel, retry failed expertise domains, and fetch the persisted records when done.
Run several LLMs on the same entity and merge the results -- conflicts are detected per field and resolved by rules or an LLM arbiter with recorded reasoning.
Chat-shaped by design: when pre-flight classification flags a mismatch ("Titan is a moon, not a planet"), the job pauses and your agent answers the question to resume -- no restart.
On claude.ai, Entity Enricher connects as a remote connector with OAuth 2.1 -- no keys to copy. The server implements dynamic client registration, PKCE, and standard discovery metadata, so the whole flow is: add the connector, approve the consent screen, done. The connection acts with your own role and can be revoked at any time from the Connected Apps tab.
In claude.ai, open Settings, then Connectors, and choose "Add custom connector".
Paste the server URL: https://entityenricher.ai/api/mcp/ (keep the trailing slash).
Sign in to Entity Enricher and approve the consent screen -- the tools appear in your chat immediately.
Clients configured by file use an API key instead: create one under Settings, then API Keys in the web app, and reference it in an X-API-Key header. The same configuration shape works for Claude Desktop, Cursor, and any MCP client that supports streamable HTTP.
{
"mcpServers": {
"entity-enricher": {
"url": "https://entityenricher.ai/api/mcp/",
"headers": { "X-API-Key": "ent_your_api_key" }
}
}
}Full client configurations, the complete tool reference, and recipe walkthroughs are in the MCP server documentation.
Pasting rows into a chat gives you unvalidated text you still have to clean up. Going through the MCP server gives your agent the full platform: schema-validated output, multi-model cross-checking, semantic IDs that dedupe entities across runs, persisted records with cost tracking, and hallucination defenses -- while the conversation stays the control surface.
Create a free account, add the connector to claude.ai or your editor, and run your first schema-validated enrichment from chat in minutes.
Get Started Free