# Entity Enricher > Entity Enricher turns any entity — a company, a drug, a property, a person, a product — into > validated, structured JSON by querying multiple LLMs against a user-defined schema. It > cross-checks models against each other (multi-model fusion), blocks wrong-entity enrichment > (pre-flight classification), defends against hallucination in 8 layers, and outputs in 40 > languages. Use it from AI agents via the hosted MCP server, from code via the REST API, or > from n8n / Make.com. Entity Enricher treats LLMs as queryable knowledge bases: it extracts facts, classifications, and relationships from model training data — optionally grounded in attached documents (PDFs, images, audio, office files) — into schema-validated output, and stores every enrichment record with cost and token analytics. Full documentation text for LLM ingestion: https://entityenricher.ai/llms-full.txt ## Connect from an AI agent (MCP) Entity Enricher exposes a hosted Model Context Protocol server that AI agents connect to directly: - MCP endpoint: https://entityenricher.ai/api/mcp/ (Streamable HTTP; keep the trailing slash) - Auth: OAuth 2.1 with dynamic client registration and PKCE — discovery at https://entityenricher.ai/.well-known/oauth-protected-resource/api/mcp — or an X-API-Key header for static-config clients - 29 tools + 2 resources: schema authoring, sample generation, single and batch enrichment, multi-model fusion, model benchmarks, async job control - [MCP Server (claude.ai / Claude Desktop / Code / Cursor)](https://entityenricher.ai/docs/integrations/mcp): Use Entity Enricher directly from claude.ai, Claude Desktop, Claude Code, Cursor, and other MCP clients. Embedded Model Context Protocol server with 29 tools, 2 resources, OAuth 2.1 or X-API-Key auth, and interactive classification-mismatch resume — the killer chat-shaped feature. - [MCP Data Enrichment - Connect Claude & AI Agents](https://entityenricher.ai/learn/mcp-data-enrichment): Enrich data from claude.ai, Claude Desktop, Claude Code, or Cursor via Entity Enricher's hosted MCP server: 29 tools for schemas, single and batch enrichment, fusion, and benchmarks. OAuth 2.1 or X-API-Key. - [Client configs & recipes (GitHub)](https://github.com/TOT-Concept/mcp-server-entity-enricher): Copy-paste client configurations, recipe walkthroughs, and the full tool reference. ## API - [API Reference](https://entityenricher.ai/docs/api): Complete REST API reference with authentication, endpoint documentation, SSE streaming, code examples in Python and curl. - [API Keys](https://entityenricher.ai/docs/platform/api-keys): Create and manage API keys for programmatic access. Organization access keys with role-based permissions, bearer tokens, and scoped access. ## Documentation - [Core Concepts](https://entityenricher.ai/docs/concepts): Understand the core concepts behind Entity Enricher: schemas as structured questions, LLM and document knowledge sources (PDF, image, audio), semantic IDs for stable identity, and quality controls. - [Enrichment Flow](https://entityenricher.ai/docs/enrichment/flow): Step-by-step walkthrough of how Entity Enricher processes a single entity: from input through classification, multi-expertise LLM calls, to structured output. - [Enrichment Strategies](https://entityenricher.ai/docs/enrichment/strategies): Compare single-pass and multi-expertise enrichment strategies. Learn when to use each approach for optimal accuracy and cost efficiency. - [Pre-flight Classification](https://entityenricher.ai/docs/enrichment/classification): How pre-flight classification prevents hallucination by verifying entity types before enrichment. Match, mismatch, ambiguous, and unknown statuses explained. - [Multi-Model Fusion](https://entityenricher.ai/docs/enrichment/fusion): How Entity Enricher fuses results from multiple AI models: conflict detection, rule-based merging, and LLM arbitration with structured reasoning. - [Multilingual Enrichment](https://entityenricher.ai/docs/enrichment/multilingual): Enrich entities in up to 40 languages simultaneously. Language-keyed JSON format compatible with PostgreSQL, MySQL, MongoDB, SQL Server, and all major databases. - [Semantic IDs](https://entityenricher.ai/docs/enrichment/semantic-ids): Give enriched objects a stable, organization-scoped semantic ID so the same real-world entity dedupes across enrichments, languages, and time. Embedding-based entity resolution with a tunable similarity threshold. - [Semantic IDs Page — Curate Your Concept Vocabulary](https://entityenricher.ai/docs/semantic-ids): Guide to the Semantic IDs page: browse concepts with slice-local similarity, curate a vocabulary with a probe-then-add loop, review near-duplicates below the threshold, inspect a concept's neighbourhood, batch-resolve a CSV, delete safely, and migrate the embedding model with no pause. - [Database Sync — Mirror Enrichments Into Your Own PostgreSQL](https://entityenricher.ai/docs/database-sync): Create a database sync on a schema and mirror enrichments as relational tables in your own PostgreSQL: downloadable SQL snapshot, idempotent delta feed with lease/acknowledge semantics, configurable database keys, admission gate, and multilingual locale flattening. - [Document Attachments](https://entityenricher.ai/docs/enrichment/attachments): Attach PDFs, images, audio recordings, Word, Excel, PowerPoint, EPUB, HTML, CSV and text files to any enrichment, schema generation, or playground request. Native PDF, vision and audio support, plus server-side extractors for office formats. - [Hallucination Prevention](https://entityenricher.ai/docs/enrichment/hallucination-prevention): How Entity Enricher prevents LLM hallucinations in structured data: 8 defense layers including pre-flight classification, multi-model consensus, nullable fields, and conflict arbitration. - [Cost Optimization & Prompt Caching](https://entityenricher.ai/docs/enrichment/cost-optimization): How Entity Enricher minimizes LLM cost: prompt caching of the shared prefix across parallel calls, per-expertise schema subsetting, schema-less text channels, cheap pre-flight classification gating, and fewer wasted retries. - [AI Schema Generation](https://entityenricher.ai/docs/schemas/generation): Generate JSON schemas from sample data using AI with self-correction. Up to 6 automatic correction attempts, 8 validation rules, and post-processing. - [Workflow Editor](https://entityenricher.ai/docs/workflow-editor): Build an enrichment end to end: sample data, a visual schema editor with drag-and-drop ordering, inline editing, keyboard navigation, and AI-assisted natural language modifications. - [Ambiguity Check](https://entityenricher.ai/docs/schemas/ambiguity): Find schema properties whose name admits more than one meaning — or none at all — see the competing readings the AI could pick, and pin each property to a single question. - [Validation Rules](https://entityenricher.ai/docs/schemas/validation): 8 validation rules that ensure schema quality: type checking, expertise validation, $ref targets, and more. Automatic LLM self-correction on failure. - [Batch Processing](https://entityenricher.ai/docs/batch): Process up to 100 entities in parallel with real-time progress tracking, automatic fusion, and export to JSON or Excel. - [Playground](https://entityenricher.ai/docs/playground): Test custom prompts with multiple AI models. View history, response formatting, token counts, and costs per execution. - [History](https://entityenricher.ai/docs/history): Browse, search, and filter enrichment records. View detailed results with integrated schema editor, grouping options, and batch operations. - [Cost Dashboard](https://entityenricher.ai/docs/cost-dashboard): Real-time cost analytics with time-series charts, performance metrics by model, language variants, and token ranges. Compare costs across time periods. - [Organizations & Roles](https://entityenricher.ai/docs/platform/organizations): Multi-tenant organization management with 4 roles: admin, owner, editor, operator. Join, leave, and manage organizations. - [Models & Pricing](https://entityenricher.ai/docs/platform/models): Manage LLM models and providers. Automatic pricing sync, model validation, health checks, and bring-your-own-key support. - [Model Routes & Capabilities](https://entityenricher.ai/docs/platform/model-routes): Why an LLM's capabilities depend on the API endpoint it is called on — Chat Completions versus the Responses API — and how Entity Enricher probes each route instead of trusting scraped claims. - [Model Benchmarks](https://entityenricher.ai/docs/platform/benchmarks): Compare LLM models on a saved, repeatable enrichment scenario — output and total cost side by side — to pick the right model for the job. - [Benchmark Scoring](https://entityenricher.ai/docs/platform/benchmark-scoring): How Entity Enricher grades benchmark results against a gold reference: a tiered equivalence ladder (exact, embedding, LLM judge), set-aligned array precision/recall/F1, and completeness/correctness sub-scores. ## Integrations - [n8n Connector](https://entityenricher.ai/docs/integrations/n8n): Install and configure the Entity Enricher n8n community node. Build automated enrichment workflows with 400+ app integrations. - [Make.com Connector](https://entityenricher.ai/docs/integrations/make): Use Entity Enricher inside Make.com (formerly Integromat) scenarios. One-call enrichment, 7 first-class modules, dynamic dropdowns, plan-limit-aware errors. Designed for the Iterator pattern instead of opaque batch modules. - [Ollama Tunnels](https://entityenricher.ai/docs/integrations/tunnels): Connect a local Ollama on your laptop to Entity Enricher with one command. Browser-OAuth pairing, in-process WebSocket multiplexer, no new public ports, instant revoke. - [ee-database Sync Client](https://entityenricher.ai/docs/integrations/ee-database): Mirror your Entity Enricher database sync into your own PostgreSQL with the open-source ee-database client. Outbound WebSocket, snapshot bootstrap, lease/ack delta feed, idempotent revision-guarded upserts, instant revoke — your connection string never leaves your machine. ## Guides - [What is Entity Enrichment? Complete Guide](https://entityenricher.ai/learn/what-is-entity-enrichment): Learn what entity enrichment is, how it works, and why AI-powered enrichment with multiple models produces more accurate structured data than traditional approaches. - [Multi-Model Enrichment & Fusion](https://entityenricher.ai/learn/multi-model-enrichment): Run multiple AI models in parallel, detect field-level conflicts, and fuse results with LLM arbitration or rule-based voting for maximum data accuracy. - [AI Schema Generation](https://entityenricher.ai/learn/schema-generation): Generate JSON schemas from sample data using AI with automatic self-correction. 8 validation rules, expertise domain detection, and natural language editing. - [Batch Processing](https://entityenricher.ai/learn/batch-processing): Enrich up to 100 entities in parallel with real-time SSE progress, automatic multi-model fusion, and export to JSON or Excel with conflict reports. ## Use cases - [Pharmaceutical Entity Enrichment](https://entityenricher.ai/use-cases/pharma): Enrich pharmaceutical entities with AI. Extract regulatory status, clinical trial phases, molecular properties, and safety profiles with schema-driven multi-model enrichment. - [Financial Entity Enrichment](https://entityenricher.ai/use-cases/finance): Enrich financial entities with AI. Extract funding rounds, market cap, risk indicators, and subsidiary structures with multi-model validation. - [Legal & Compliance Entity Enrichment](https://entityenricher.ai/use-cases/legal): Enrich legal entities with AI. Extract jurisdiction data, compliance status, corporate structure, and regulatory filings with structured output. - [Academic Research Entity Enrichment](https://entityenricher.ai/use-cases/research): Enrich research entities with AI. Extract citation counts, h-index, institutional affiliations, methodology details, and publication metadata. - [Real Estate Entity Enrichment](https://entityenricher.ai/use-cases/real-estate): Enrich real estate entities with AI. Extract zoning data, market valuation, neighborhood demographics, permit history, and property characteristics. - [E-commerce Product Data Enrichment](https://entityenricher.ai/use-cases/ecommerce): Enrich product catalogs with AI. Complete attributes, categories, compliance data, and multilingual descriptions from supplier feeds and spec-sheet PDFs -- deduplicated with semantic IDs. ## Comparisons - [Entity Enricher vs Parallel AI](https://entityenricher.ai/compare/parallel-ai): How Entity Enricher compares to Parallel AI's Task API. Multi-model fusion, your documents as sources, semantic-ID dedup, and sync to your own PostgreSQL vs a single web-research agent with per-field confidence. - [Entity Enricher vs AI Research Agents](https://entityenricher.ai/compare/ai-research-agents): How Entity Enricher compares to spreadsheet AI research agents like Claygent, Datablist, and PromptLoop. Typed nested schemas, multi-model fusion, entity dedup, and relational database sync vs one prompt per column. - [Entity Enricher vs a DIY LLM Pipeline - Buy vs Build](https://entityenricher.ai/compare/diy-llm-pipeline): How Entity Enricher compares to building your own pipeline with Instructor, BAML, PydanticAI, or LangChain. Multi-model fusion, arbitration, semantic IDs, and batch — without the maintenance. - [Entity Enricher vs LlamaExtract](https://entityenricher.ai/compare/llama-extract): How Entity Enricher compares to LlamaExtract. Enrich from LLM knowledge, web search, and documents with multi-model arbitration and semantic IDs vs single-pass document extraction. - [Entity Enricher vs Diffbot](https://entityenricher.ai/compare/diffbot): How Entity Enricher compares to Diffbot. Enrich the entities you define against a custom schema with multi-model arbitration vs querying a prebuilt web knowledge graph. - [Entity Enricher vs Supabase MCP — JSON to Relational Data](https://entityenricher.ai/compare/supabase-mcp): See how a misplaced supplier price corrupts a product catalog. Compare Supabase MCP with EE's reviewed relationships, indexes and database sync. ## Optional - [LLM Data Extraction & AI Data Enrichment Features](https://entityenricher.ai/features): LLM data extraction features: AI schema generation, multi-expertise enrichment, multi-model fusion, batch processing, cost analytics, and REST API. Everything for schema-driven AI data enrichment. - [AI Data Enrichment Pricing - Free, Pro & Premium Plans](https://entityenricher.ai/pricing): AI data enrichment pricing: free tier, pay-per-use Pro plan, and unlimited Premium at $99/month. Usage-based credits, no markup with your own AI provider API keys, multilingual enrichment included on every plan. - [LLM models compared: pricing, capabilities, benchmarks](https://entityenricher.ai/llm-models): Every model of the Entity Enricher catalog with its vendor pricing, measured capabilities and benchmark scores for structured extraction — sortable, filterable, and downloadable as JSON. - [Changelog](https://entityenricher.ai/changelog): New features and improvements to Entity Enricher: MCP and claude.ai connectivity, model benchmarks, semantic IDs, document attachments, multilingual UI, and more. - [Terms of Service](https://entityenricher.ai/terms): Terms of Service for Entity Enricher, operated by TOT Concept SAS. Covers usage, pricing, data retention, and GDPR compliance. - [Privacy Policy](https://entityenricher.ai/privacy): Privacy policy for Entity Enricher. Learn how we handle your data, GDPR rights, and data protection measures.