mcpField
S
Senzing

Senzing

Remote
senzing.comDeveloper Tools0Updated today

About Senzing

Senzing MCP Server supports entity resolution and provides grounded guidance on Senzing documentation, SDKs, and troubleshooting. It is for teams mapping and linking records across sources, with local data processing and explanations for why records match or do not match.

Key features

  1. 01Guide source-data mapping
  2. 02Analyze and link entity records
  3. 03Generate SDK scaffolding and references
  4. 04Search docs, examples, and error fixes
  5. 05Provide workflow prompts

Use cases

  1. 01Deduplicate customer or payment records
  2. 02Map data from multiple source files
  3. 03Build an SDK integration
  4. 04Investigate Senzing error codes

MCP support

The hosted remote MCP server exposes tools for data mapping, entity analysis, SDK development, documentation search, troubleshooting, and reporting, along with workflow prompts. Mapping and resolution run locally in the user's environment; the server requires no authentication.

Configuration

Remote servers need no local install. Pick your client, then copy the config or install in one click.

Endpointhttps://mcp.senzing.com/mcp
Copy install command
Claude CodeTerminal
Cursor.cursor/mcp.json
VS Code.vscode/mcp.json
Claude Desktopclaude_desktop_config.json
Windsurf~/.codeium/windsurf/mcp_config.json

Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.

Tools

13
ParameterTypeRequiredDescription
versionstringNoSenzing version (default: "current").
file_pathsarray|nullNoFile paths to analyze (Senzing JSON or JSONL files). Commands will be generated for each path.
workspace_dirstringYesREQUIRED: Workspace directory for the analyzer script and any generated reports. Must be a writable absolute or relative path that already exists in your environment. Do NOT assume `/tmp` exists — some environments (e.g. Kiro) do not provide it. Typical values: Linux: `/tmp` or `~/sz-workspace`; macOS: `~/sz-workspace`; Kiro/sandboxed: an explicit path under `~`, e.g. `~/sz-workspace`. Declared as a non-optional `String` (no `#[serde(default)]`) so it appears in the generated JSON `inputSchema` `required` array. The prose has always said REQUIRED, but as an `Option` it was schema-optional, so schema-respecting clients omitted it and looped on the `needs_param` guidance. An empty string still deserializes; the handler catches that and returns the guidance as an `isError` result.

Details

From the project README

Senzing

Senzing MCP Server

Entity resolution for Identity Intelligence: dedupe, match, and link people and company records across systems. Know who is who. Reveal hidden relationships.

MCP Registry: com.senzing/mcp Transport: Streamable HTTP Authentication: none Hosted server: free to use

Install · What you can ask · Tools · Prompts · Privacy and terms · Questions and feedback


The Senzing® MCP server is a forward-deployed engineer in a box: ask it anything about Senzing. Every answer is grounded in Senzing's own documentation and indexed code rather than model training data. Use it to deduplicate lists, match or merge records across sources, screen against watchlists, and produce the resolved entities that golden records for MDM and customer 360 are built from.

Because it carries the full runbook, you don't drive it one step at a time. Hand your AI assistant a few unmapped data files and describe what you want, and it can map your sources, deploy and load Senzing, resolve and analyze the data, and build an interactive explorer over the results. What is normally a multi-week integration becomes an afternoon, with no entity-resolution background required.

You just ask. Then start answering who is who and who is related to whom, or build an entity-resolved knowledge graph that surfaces the connections nobody declared. Every resolution carries its evidence: why two records came together, or why they didn't. Your assistant can surface that explanation for any match you question.

Install

Endpoint: https://mcp.senzing.com/mcp (Streamable HTTP, no authentication) · Registry name: com.senzing/mcp · The hosted server is free to use; Senzing software is licensed separately.

Add to Claude Install in VS Code Install in VS Code Insiders Install in Visual Studio Install in Cursor

The buttons above add the server in one click (Claude through its Connectors Directory). For other clients:

ClientHow to connect
Claude (claude.ai, Desktop, Cowork)Install from the Connectors Directory, or Customize → Connectors → + Add → Add custom connector with the URL https://mcp.senzing.com/mcp (Team and Enterprise owners: Organization settings → Connectors)
Claude Codeclaude mcp add --transport http senzing https://mcp.senzing.com/mcp
OpenAI Codexcodex mcp add senzing --url https://mcp.senzing.com/mcp
Gemini CLIgemini mcp add --transport http senzing https://mcp.senzing.com/mcp
VS Code (manual).vscode/mcp.json — see below
Cursor (manual).cursor/mcp.json or ~/.cursor/mcp.json — see below
Kiro.kiro/settings/mcp.json or ~/.kiro/settings/mcp.json — see below
stdio-only clientsRun it through mcp-remote — see below
Any other MCP clientAdd a remote (Streamable HTTP) server with the URL https://mcp.senzing.com/mcp
VS Code — .vscode/mcp.json
{
  "servers": {
    "senzing": {
      "type": "http",
      "url": "https://mcp.senzing.com/mcp"
    }
  }
}
Cursor and Kiro — mcp.json
{
  "mcpServers": {
    "senzing": {
      "url": "https://mcp.senzing.com/mcp"
    }
  }
}
stdio-only clients — via mcp-remote
{
  "mcpServers": {
    "senzing": {
      "command": "npx",
      "args": ["-y", "mcp-remote", "https://mcp.senzing.com/mcp"]
    }
  }
}
Codex — ~/.codex/config.toml
[mcp_servers.senzing]
url = "https://mcp.senzing.com/mcp"

Tip: for the best experience, use an agentic coding client (Claude Code, Codex, Gemini CLI, Cursor); if your environment restricts network access, allowlist mcp.senzing.com. Mapping your own data and running Senzing need an assistant that can run code; questions, documentation search and SDK code work in any client.

What you can ask

  • "Match these payment records to my customers and show me which ones didn't match."
  • "Dedupe this list of people and show me the duplicates it found."
  • "Here are three customer files from different systems. Build me a demo that finds the same people across them and lets me explore the matches."
  • "Build me a service that matches incoming payment records to my customer master as they arrive."
  • "How would entity resolution help with my problem, and what would it cost?"

Tools

Each piece is also available on its own: SDK code in Python, Java and C# (official SDKs) and Rust and TypeScript/Node.js (community SDKs), installation and deployment, error troubleshooting, V3→V4 migration, and pricing and ROI.

AreaToolWhat it does
Start hereget_capabilitiesOverview of the server and suggested workflows
Map your datamapping_workflowGuided 8-step mapping of source data to the Senzing Entity Specification
analyze_recordAnalyzer script that validates your mapped records on your machine
download_resourceWorkflow resources, as a fallback when a URL can't be fetched
Build with the SDKsdk_guideInstall and setup across Linux (apt/yum), macOS, Windows and Docker, for each SDK language
generate_scaffoldSDK code from real indexed snippets: Python, Java, C# (official); Rust, TypeScript/Node.js (community)
get_sdk_referenceMethod signatures per language binding, flags, response schemas, V3→V4 migration
Find answerssearch_docsFull-text search of Senzing documentation
find_examplesCode examples from the indexed Senzing GitHub repositories
explain_error_codeSenzing error codes with causes and fixes
Try itget_sample_dataReal sample data: CORD datasets (Las Vegas, London, Moscow) and the Senzing truth set
Reportreporting_guideReporting and visualization patterns: export, SQL analytics, data mart, dashboards, graph
Feedbacksubmit_feedbackSend feedback or request a free evaluation license

Prompts

The server also ships 31 ready-made prompts: workflow templates your client can offer as slash commands or menu items. A few of them:

PromptWhat it walks through
why-senzingExplains Senzing for a given audience
plan-a-pocPlanning a proof of concept
map-data-sourceMapping a data source to the Entity Specification
install-senzingInstalling and configuring the SDK
build-sdk-integrationBuilding an SDK integration in your language
troubleshoot-errorDiagnosing a Senzing error code
migrate-v3-to-v4Migrating from Senzing V3 to V4

Ask your assistant to list the Senzing prompts to see them all.

More from Senzing

ProjectWhat it isWhere
Claude Code pluginSenzing skills for Claude Code: install, analyze, build, demo, troubleshoot, recipes and moreSenzing/senzing-claude-plugins
Agent SkillPortable skill that teaches any skill-capable agent how to use this serverSenzing/senzing-mcp-skill
Ask SenzingChat with the Senzing docs in your browser, powered by this serverask.senzing.com
Building your own agentHow to keep an API-driven agent's answers grounded in this serverdocs/grounded-agents.md

Privacy and terms

The hosted server has its own privacy policy. In short:

  • The server needs no live Senzing instance and no authentication. Your data files never leave your machine: mapping and resolution run locally, and only the tool arguments you send (field names, settings, questions) reach this server.
  • Senzing is a headless developer library your assistant deploys locally in your environment. Mapping and resolution run there, never on Senzing's servers, and Senzing contains only the data you load into it.
  • Do not send personal data, credentials or other sensitive information in prompts or tool arguments. Tool calls and their parameters may be logged.
  • The hosted server is provided AS-IS, without warranty or support commitment. The server also exposes these terms as the senzing://terms-of-service resource.

Questions and feedback

Use senzing.com/contact or [email protected], or ask your assistant to use submit_feedback.

See LICENSE.

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.