Claude Code

\n\n```bash\nclaude mcp add --transport http entity-enricher https://entityenricher.ai/api/mcp/\n```\n\nThen run `/mcp` in a session and pick **Authenticate** — your browser opens the consent page.\nMore options (project `.mcp.json`, API-key fallback): [examples/claude-code/](https://github.com/TOT-Concept/mcp-server-entity-enricher/blob/main/examples/claude-code/)\n

\n\n

claude.ai

\n\n**Settings → Connectors → Add custom connector** with URL\n`https://entityenricher.ai/api/mcp/`, then click **Authorize** on the consent screen.\nWalkthrough: [examples/claude-ai-remote.md](https://github.com/TOT-Concept/mcp-server-entity-enricher/blob/main/examples/claude-ai-remote.md)\n

\n\n

Cursor / other OAuth-capable clients

\n\nRegister the URL with no headers and the client prompts you to sign in:\n[examples/cursor/mcp.json](https://github.com/TOT-Concept/mcp-server-entity-enricher/blob/main/examples/cursor/mcp.json)\n

\n\n### Option 2 — API key (static JSON configuration)\n\nFor clients configured via a JSON file rather than an interactive sign-in (Claude Desktop,\nContinue, Zed) — and for headless/CI use.\n\n1. In the [Entity Enricher web UI](https://entityenricher.ai): **Settings → API Keys → New\n organization access key**. Pick a role — operator (read-mostly), editor (create/edit\n schemas), or owner (full control, required for benchmarks). Copy the `ent_…` value; it's\n only shown once.\n2. For **Claude Desktop**, edit `~/Library/Application Support/Claude/claude_desktop_config.json`\n (macOS) or `%APPDATA%\\Claude\\claude_desktop_config.json` (Windows):\n\n ```json\n {\n \"mcpServers\": {\n \"entity-enricher\": {\n \"url\": \"https://entityenricher.ai/api/mcp/\",\n \"headers\": { \"X-API-Key\": \"ent_your_key_here\" }\n }\n }\n }\n ```\n\n Restart Claude Desktop. Full file: [examples/claude-desktop/](https://github.com/TOT-Concept/mcp-server-entity-enricher/blob/main/examples/claude-desktop/)\n\n### Try it\n\n> List my Entity Enricher schemas, then enrich \"Sanofi\" against the pharmaceutical company\n> schema in English and French.\n\nClaude discovers the tools automatically, confirms the model and schema choice with you, and\nreturns the structured result inline.\n\n## Examples & recipes\n\nClient configs and copy-paste chat walkthroughs live in [examples/](https://github.com/TOT-Concept/mcp-server-entity-enricher/blob/main/examples/):\n\n| Recipe | What it covers |\n|---|---|\n| [Schema from sample](https://github.com/TOT-Concept/mcp-server-entity-enricher/blob/main/examples/recipes/schema-from-sample.md) | generate a sample → schema → refine → first enrichment |\n| [Database sync](https://github.com/TOT-Concept/mcp-server-entity-enricher/blob/main/examples/recipes/database-sync.md) | schema → designed tables → publish → pair `ee-database` → migrations |\n| [Batch enrichment](https://github.com/TOT-Concept/mcp-server-entity-enricher/blob/main/examples/recipes/batch-enrichment.md) | entity lists, external APIs, async polling, partial-failure retry |\n| [Model benchmark](https://github.com/TOT-Concept/mcp-server-entity-enricher/blob/main/examples/recipes/model-benchmark.md) | scenarios, gold references, auto-scored model comparison |\n\nPer-client setup and examples: [Claude Code](https://github.com/TOT-Concept/mcp-server-entity-enricher/blob/main/examples/claude-code/) · [claude.ai](https://github.com/TOT-Concept/mcp-server-entity-enricher/blob/main/examples/claude-ai-remote.md) · [Claude Desktop](https://github.com/TOT-Concept/mcp-server-entity-enricher/blob/main/examples/claude-desktop/) · [Cursor](https://github.com/TOT-Concept/mcp-server-entity-enricher/blob/main/examples/cursor/mcp.json)\n\n## Tools\n\n\n**57 tools**, spanning the full schema-authoring and enrichment surface:\n\n| Category | Tool | Description |\n|---|---|---|\n| Discovery | `list_models` | List the LLM models, languages, strategies, and (when the org has a plan with limits) the operational profile_limits available to the caller. |\n| Schemas | `generate_sample` | Generate a realistic sample entity JSON from a free-text request — the entry point of the schema-authoring loop. |\n| Schemas | `list_schemas` | List saved JSON schemas in your organization, pinned ones first. |\n| Schemas | `get_schema` | Fetch the full content of a saved schema by ID, including all properties, identifying fields, expertise domains, and validation rules. |\n| Schemas | `create_schema_from_sample` | Generate and auto-save a JSON schema whose paths and types strictly follow an approved sample. |\n| Schemas | `save_schema` | Persist a schema you authored directly (no LLM call, no cost) as a new saved schema. |\n| Schemas | `update_schema` | Update a saved schema without an LLM call: rename, replace the schema_content, change tags, pin/unpin, or toggle the ambiguity check. |\n| Schemas | `get_schema_part` | Read a part of a saved schema without fetching the whole document. |\n| Schemas | `get_enum_candidates` | Out-of-set values enrichment has returned for each OPEN enum of a saved schema (closed: false, the generation default), with counts — read from the schema's recent records at call… |\n| Schemas | `update_schema_property` | Edit ONE property of a saved schema's working copy by path — rename, change type or $ref, description, examples, flags — or remove it, without sending the full schema_content. |\n| Schemas | `add_schema_property` | Add a property to an object of a saved schema's working copy: a scalar, an inline nested object (optionally with sub-properties, or untyped to fill later), or a $ref to an… |\n| Schemas | `move_schema_property` | Move ONE property of a saved schema's working copy into another container — a $defs entity ('$defs.X'), an inline object (object path, trailing '[]' enters an array's item type),… |\n| Schemas | `resolve_unify_proposal` | Resolve ONE cross-site unification proposal on a saved schema (x-entityMap.proposals; issue #181). |\n| Schemas | `publish_schema` | Publish a linked schema's working copy as its contract (publish model): enrichment and the linked database syncs follow the published content only, so structural edits (new… |\n| Schemas | `delete_schema` | Soft-delete a saved schema by ID (restorable server-side shortly after; permanent deletion stays in the web UI). |\n| Schemas | `analyze_sample` | Analyze a sample entity before schema generation: ambiguity and identity scoping, two parallel model calls behind one request. |\n| Schemas | `analyze_schema` | Analyze a saved schema with the same two checks as analyze_sample — ambiguity and identity scoping — and write the verdicts onto its properties. |\n| Enrichment & fusion | `start_batch_enrichment` | Start an asynchronous batch enrichment against a JSON schema and return {job_id, total} immediately. |\n| Enrichment & fusion | `fetch_entities` | Fetch a JSON array of entities from an external REST API (GET), server-side — the input step before start_batch_enrichment. |\n| Enrichment & fusion | `enrich_entity` | Run a multi-model enrichment of a single entity against a JSON schema, returning the fused/best structured result. |\n| Enrichment & fusion | `retry_expertises` | Re-run only the FAILED expertise domains of an existing multi-expertise enrichment record, merging the recovered values back into the record — no re-payment for the domains that… |\n| Enrichment & fusion | `merge_records` | Merge 2+ enrichment records of the same entity into one fused result — the manual / re-run counterpart of the automatic fusion that follows a 2+ model enrich_entity or batch run. |\n| Job control | `get_job_status` | Poll the status of an asynchronous LLM job — the middle step of every start → poll → fetch flow (start_batch_enrichment, generate_sample, run_benchmark, retry_expertises). |\n| Job control | `cancel_job` | Cancel a pending, running, or paused LLM job started by start_batch_enrichment, generate_sample, run_benchmark, or retry_expertises. |\n| Job control | `answer_job_question` | Answer the clarification questions of a paused job and resume it — the reply half of the interactive loop used by generate_sample's document-grounded planner (get_job_status… |\n| Records & stats | `list_records` | List past enrichment records in your organization, most recent first. |\n| Records & stats | `get_record` | Fetch a single enrichment record by ID, including the full structured output, validation errors, per-expertise verdicts and metrics. |\n| Records & stats | `get_stats` | Aggregated statistics over your organization's enrichment records: totals, success rate, token usage, and cost summary. |\n| Benchmarks | `list_benchmark_scenarios` | List the organization's benchmark scenarios (saved, reusable enrichment tests: schema + entity + strategy + scoring config). |\n| Benchmarks | `get_benchmark_scenario` | Fetch one benchmark scenario with its per-model results (quality / cost / speed scores; results whose config_hash differs from the scenario's are stale — re-run those models). |\n| Benchmarks | `get_benchmark_scenario_results` | Filter, rank and cap a scenario's per-model results — the same rows get_benchmark_scenario returns, narrowed to what you asked for. |\n| Benchmarks | `create_benchmark_scenario` | Create a benchmark scenario — a reusable model test. |\n| Benchmarks | `update_benchmark_scenario` | Update a benchmark scenario. |\n| Benchmarks | `set_benchmark_reference` | Save a scenario's gold reference — the expected output each model result is scored against, and the gate between create_benchmark_scenario and run_benchmark. |\n| Benchmarks | `delete_benchmark_scenario` | Delete a benchmark scenario and its results. |\n| Benchmarks | `run_benchmark` | Launch a benchmark run — the final step of the benchmark lifecycle: execute the scenario's task (enrichment / sample generation / schema generation) with each selected model… |\n| Attachments | `upload_attachment` | Upload a file (base64-encoded) so it can be used as source material in LLM flows. |\n| Attachments | `delete_attachment` | Permanently remove an attachment from the server by id. |\n| Database Sync | `list_database_syncs` | List the database syncs registered on a saved schema, with pending delta counts. |\n| Database Sync | `list_entity_states` | Browse the current entity state of a schema — the deduplicated, last-write-wins merged rows the entity layer holds (and every linked database mirrors), NOT the per-run records of… |\n| Database Sync | `create_database_sync` | Connect a database to a saved schema — the opt-in that turns enrichments into relational SQL deltas the user applies to their own PostgreSQL/MySQL/SQLite with the ee-database CLI… |\n| Database Sync | `assign_sync_host` | Assign (or clear) the sync host that provisions a database sync in managed ee-database mode: the assigned host claims the credential, creates the physical database if missing and… |\n| Database Sync | `classify_database_model` | Re-run the database-model classification pass on a saved schema: an LLM proposes each property's SQL contract — database_key (identity), the index intent ('search' for text a… |\n| Database Sync | `delete_database_sync` | Delete a database sync and its queued deltas. |\n| Database Sync | `create_database_credential` | (Re)issue the sync-client credential of a database sync — the pairing step of the ee-database CLI workflow. |\n| Database Sync | `fetch_database_deltas` | Fetch the next FIFO window of SQL deltas for a database sync. |\n| Database Sync | `ack_database_deltas` | Acknowledge applied database deltas up to an id: releases the lease and, per the database's options, purges delivered copies and fully-delivered entity state. |\n| Database Sync | `sync_records_to_database` | Push already-stored enrichment output into the entity layer, so it reaches the schema's database sync. |\n| Semantic IDs | `list_semantic_concepts` | Browse your organization's semantic-ID vocabulary — the org-scoped concepts that near-duplicate objects resolve to and database syncs key on. |\n| Semantic IDs | `get_semantic_concept` | One concept in full: its surface-form aliases (the texts that resolve to it, one canonical), the identity source keys it was composed from, the records that resolved to it, and… |\n| Semantic IDs | `probe_semantic_concept` | Dry-run the resolution ladder for a text against one concept type — what an enrichment would do with it — without creating or bumping anything. |\n| Semantic IDs | `add_semantic_concept` | Add a concept to the vocabulary at usage 0 (editor role) — or, with alias_of, add the text as a new SURFACE FORM of that existing concept instead of a concept of its own. |\n| Semantic IDs | `update_concept_alias` | Manage one surface form (alias) of a concept (editor role): action='remove' prunes a captured variant so its text stops resolving to the concept — the group's LAST surface form is… |\n| Semantic IDs | `import_semantic_concepts` | Resolve a batch of identity texts (up to 1000) against one concept type through the same ladder an enrichment uses. |\n| Semantic IDs | `merge_semantic_concepts` | Fold one concept (the loser) into another (the winner) — the resolution of a duplicates-band pair. |\n| Semantic IDs | `delete_semantic_concepts` | Delete concepts by explicit ids, whole concept types, or every unused concept of a scope. |\n| Semantic IDs | `migrate_semantic_embeddings` | The org's embedding-model migration — the only sanctioned way to move existing concepts between embedding models. |\n\n\nTool behaviour is identical to the REST endpoints they wrap — same validation, billing and plan\nlimits as the web app. Write tools require the **editor** role; benchmark tools require\n**owner** plus a plan that includes Model Benchmarks.\n\n### Resources\n\nResources let the client *browse* data without a tool call — both render as Markdown.\n\n| Resource | URI template |\n|---|---|\n| Saved schema | `enricher://schemas/{schema_id}` |\n| Enrichment record | `enricher://records/{record_id}` |\n\n## The async job pattern\n\nMCP tools can't stream, so long-running work is split into **start → poll → fetch**:\n\n1. A start tool (`start_batch_enrichment`, `generate_sample`, `run_benchmark`,\n `retry_expertises`) returns a `job_id` immediately.\n2. `get_job_status(job_id)` polls progress; paused jobs carry clarification questions that\n `answer_job_question` resolves; `cancel_job` aborts.\n3. Persisted outputs are fetched with `list_records(job_id=…)` (or the feature's own read\n tool, e.g. `get_benchmark_scenario`).\n\nJobs are held in a bounded in-memory manager — an unknown `job_id` means the job finished long\nago; go straight to the records.\n\n## Interactive classification resume\n\nThe feature that only an interactive client unlocks. With a classification model enabled, a\npre-flight check verifies the entity matches the schema type. On a mismatch the tool returns a\n**non-error** response instead of failing:\n\n```json\n{\n \"success\": false,\n \"error_code\": \"classification_warning\",\n \"message\": \"Pre-flight classification rejected the entity. ...\",\n \"classification\": {\n \"status\": \"mismatch\",\n \"reasoning\": \"Titan is a moon of Saturn, not a planet.\",\n \"confidence\": 0.97\n },\n \"job_id\": \"...\"\n}\n```\n\nClaude surfaces the reasoning, asks you to confirm, and retries with\n`force_after_classification_warning=true`. Workflow connectors (n8n, Make) have to auto-cancel\nhere — a chat can just ask.\n\n## Error codes\n\nErrors are structured dicts with an `error_code` field the client can pattern-match on:\n\n| `error_code` | When |\n|---|---|\n| `invalid_request` | Malformed UUID, mutually exclusive args, body validation failure. |\n| `prompt_limit_reached` | Daily/weekly/monthly prompt quota exhausted (HTTP 402), with period + usage details. |\n| `insufficient_credits` | Credit balance too low to start the job (HTTP 402), with balance + purchase URL. |\n| `model_limit_exceeded` / `language_limit_exceeded` | More models/languages requested than the plan allows (HTTP 402). |\n| `concurrent_job_limit_reached` | Too many active jobs for the org — wait or upgrade. |\n| `classification_warning` | ⚡ Non-error: pre-flight classifier rejected the entity (see above). |\n| `benchmarks_not_in_plan` | Benchmark tools need the owner role + a plan with Model Benchmarks (HTTP 403). |\n| `enrichment_timeout` / `schema_generation_timeout` | Job exceeded its timeout — try fewer models. |\n| `schema_generation_failed` | Upstream LLM error (HTTP 502). |\n| `cancelled` | Job cancelled mid-run (HTTP 499). |\n| `not_found` | Schema or record ID doesn't exist in your org. |\n\n## Authentication details\n\n- **OAuth 2.1 (recommended)** — any MCP client implementing the standard auth spec (claude.ai,\n Claude Code, Cursor, MCP Inspector) discovers it automatically: standard discovery via\n `/.well-known/oauth-protected-resource`, dynamic client registration, PKCE, browser consent.\n No key to create or paste. Tokens are audience-bound and instantly revocable under\n **Settings → API Keys → Connected Apps**.\n- **X-API-Key** — for clients configured via a static JSON file: `ent_…` organization access\n keys, created in Settings → API Keys. The role attached to the key (operator / editor /\n owner) gates which tools succeed.\n\n## Links\n\n- [Entity Enricher](https://entityenricher.ai) — the platform.\n- [MCP server docs](https://entityenricher.ai/docs/integrations/mcp) — the always-current web version of this guide.\n- [REST API reference](https://entityenricher.ai/docs/api) — the endpoints these tools wrap.\n- [Model Context Protocol](https://modelcontextprotocol.io) — the open spec.\n\n## About this repository\n\nThis repo contains the public documentation and client examples for the Entity Enricher MCP\nserver. The server itself is embedded in the Entity Enricher platform and maintained in the\nmain (private) monorepo; this repo is synced from it as a git subtree. Issues and discussions\nare welcome here.\n\nLicensed under the [MIT License](https://github.com/TOT-Concept/mcp-server-entity-enricher/blob/main/LICENSE).\n","changelogMarkdown","videoKey","requiredSecretsJson","[{\"name\":\"X_API_KEY\",\"scope\":\"per-user\",\"header\":\"X-API-Key\"}]","publishedAt","priceCents","trialDays","iconType","image","upstream",{"_113":114,"_115":116,"_117":-5,"_118":119,"_120":121,"_122":123},"requiredSecrets",[106],"tags",[101,102,103,104,105],"assets",[],"stats",{"_99":59,"_100":59},"ratingAvg","reviewCount","ai-agents","claude","data-enrichment","entity-enrichment","llm",{"_107":108,"_109":110,"_111":112},"name","X_API_KEY","scope","per-user","header","X-API-Key","topics",[101,102,103,104,105,124,125,126],"license","MIT","language","registryName","ai.entityenricher/enricher","homepage","https://entityenricher.ai/docs/integrations/mcp","packages",[],"mcp","mcp-server","model-context-protocol","ENV",{"_129":130,"_131":132,"_133":134},"ENVIRONMENT","production","TURNSTILE_SITE_KEY","0x4AAAAAAB8-rEI2gNDf_QhU","SENTRY_DSN","https://45fd1229f78089d5c3ee78be128030e1@o4510583261757440.ingest.de.sentry.io/4510584253448272"]