[{"_1":2,"_3":-5,"_4":-5},"loaderData",{"_5":6,"_7":8,"_9":10},"actionData","errors","root",{"_125":126},"routes/modules.$creator.$moduleId",{"_14":15,"_16":17,"_18":19,"_20":21},"routes/modules.$creator.$moduleId.endpoints",{"_11":12,"_13":-5},"error","Module '@benseverndev-oss/goldenmatch' is not configured","tools","module",{"_20":22,"_23":24,"_25":26,"_27":28,"_29":30,"_31":32,"_33":26,"_34":-5,"_35":36,"_37":-5,"_38":-5,"_39":40,"_41":42,"_43":44,"_45":46,"_47":48,"_49":50,"_51":52,"_53":54,"_55":48,"_56":57,"_58":59,"_60":-5,"_61":-5,"_62":-5,"_63":-5,"_64":65,"_66":67,"_68":69,"_70":71,"_72":73,"_74":-5,"_75":-5,"_76":-5,"_77":48,"_78":-5,"_79":80,"_81":82,"_83":84,"_85":86,"_87":88,"_89":90,"_91":92},"apiUrl","https://odel.app/api","creator","benseverndev-oss","moduleId","goldenmatch","@benseverndev-oss/goldenmatch","displayName","GoldenMatch","tagline","Find duplicate records in 30 seconds. Zero-config entity resolution, 97.2% F1 out of the box.","iconKey","https://avatars.githubusercontent.com/u/285030165?v=4","category","data","status","published","metaDescription","metaKeywords","deploymentType","external","workerName","bindingName","mcpConfig","{\"url\":\"https://goldenmatch-mcp-production.up.railway.app/mcp/\",\"transport\":\"streamable-http\",\"authType\":\"none\"}","pricingModel","free","creatorUserId","system:external-import","createdAt",1781363396000,"updatedAt",1781499599000,"sourceRegistry","official-mcp-registry","sourceUrl","https://registry.modelcontextprotocol.io/v0/servers?search=io.github.benseverndev-oss%2Fgoldenmatch","sourceVerified",true,"importedAt","upstreamStars",110,"upstreamUpdatedAt",1781492039000,"detectedTransport","detectedAuthType","probeStatus","lastProbedAt","githubRepoUrl","https://github.com/benseverndev-oss/goldenmatch","githubRepoFullName","benseverndev-oss/goldenmatch","releaseId","ext-benseverndev-oss-goldenmatch","version","2.0.0","descriptionMarkdown","\n

\n\n# 🟡 Golden Suite\n\n**A polyglot data-quality and entity-resolution toolkit. Polished, opinionated, AI-native.**\n\n*GoldenCheck profiles → GoldenFlow standardizes → GoldenMatch deduplicates → GoldenAnalysis reports, all orchestrated by GoldenPipe. With InferMap for schema mapping, a Rust extension layer for Postgres / DuckDB, and optional WebAssembly acceleration behind the edge-safe TypeScript ports.*\n\n**⚡ GoldenMatch scales from a CSV on your laptop to 100M+ rows on a Ray cluster — verified: 100,000,000 records deduped recall-complete (correct across any partitioning) in 9.2 min, with a 0.36 GB driver footprint.**\n\n
\n\n\n[![PyPI — goldenmatch](https://img.shields.io/pypi/v/goldenmatch?color=d4a017&label=pypi%3Agoldenmatch&logo=pypi&logoColor=white)](https://pypi.org/project/goldenmatch/)\n[![npm — goldenmatch](https://img.shields.io/npm/v/goldenmatch?color=cb3837&label=npm%3Agoldenmatch&logo=npm&logoColor=white)](https://www.npmjs.com/package/goldenmatch)\n[![Python](https://img.shields.io/badge/python-3.11%2B-3776ab?logo=python&logoColor=white)](https://python.org)\n[![Node](https://img.shields.io/badge/node-%3E%3D20-5fa04e?logo=nodedotjs&logoColor=white)](https://nodejs.org)\n[![License: MIT](https://img.shields.io/badge/license-MIT-green)](LICENSE)\n\n\n[![CI](https://github.com/benseverndev-oss/goldenmatch/actions/workflows/ci.yml/badge.svg)](https://github.com/benseverndev-oss/goldenmatch/actions/workflows/ci.yml)\n[![codecov](https://codecov.io/gh/benseverndev-oss/goldenmatch/graph/badge.svg)](https://codecov.io/gh/benseverndev-oss/goldenmatch)\n[![OpenSSF Scorecard](https://api.scorecard.dev/projects/github.com/benseverndev-oss/goldenmatch/badge)](https://scorecard.dev/viewer/?uri=github.com/benseverndev-oss/goldenmatch)\n[![Fellegi-Sunter beats hand-rolled Splink](https://img.shields.io/badge/Fellegi--Sunter-beats%20hand--rolled%20Splink-d4a017)](docs/benchmarks/2026-06-09-splink-bakeoff.md)\n[![DBLP-ACM F1](https://img.shields.io/badge/DBLP--ACM%20F1-96.4%25-d4a017)](packages/python/goldenmatch/README.md#benchmarks)\n\n\n[![PyPI downloads (suite)](https://img.shields.io/endpoint?url=https%3A%2F%2Fraw.githubusercontent.com%2Fbenseverndev-oss%2Fgoldenmatch%2Fbadges%2Fpypi-downloads.json)](https://pepy.tech/projects?q=goldenmatch+goldencheck+goldenpipe+goldenflow+infermap+goldencheck-types+goldensuite-mcp+goldenmatch-duckdb+goldenmatch-native+goldenmatch-embed)\n[![npm downloads (suite)](https://img.shields.io/endpoint?url=https%3A%2F%2Fraw.githubusercontent.com%2Fbenseverndev-oss%2Fgoldenmatch%2Fbadges%2Fnpm-downloads.json)](https://www.npmjs.com/~benzsevern)\n[![GitHub stars](https://img.shields.io/github/stars/benseverndev-oss/goldenmatch?style=flat&color=d4a017&logo=github)](https://github.com/benseverndev-oss/goldenmatch/stargazers)\n\n\n[![Docs](https://img.shields.io/badge/docs-docs.bensevern.dev-d4a017)](https://docs.bensevern.dev/)\n[![Wiki](https://img.shields.io/badge/wiki-github-d4a017)](https://github.com/benseverndev-oss/goldenmatch/wiki)\n[![Web UI](https://img.shields.io/badge/web%20ui-FastAPI%20%2B%20React-d4a017?logo=react&logoColor=white)](https://github.com/benseverndev-oss/goldenmatch/wiki/Web-UI)\n[![Smithery MCP](https://img.shields.io/badge/MCP-smithery-6e40c9)](https://smithery.ai/servers/benzsevern/goldenmatch)\n\n\n[![Last commit](https://img.shields.io/github/last-commit/benseverndev-oss/goldenmatch?color=d4a017&label=last%20commit)](https://github.com/benseverndev-oss/goldenmatch/commits/main)\n\n

\n\n[![GoldenMatch web workbench — pair drilldown with NL prose](https://raw.githubusercontent.com/benseverndev-oss/goldenmatch/main/packages/python/goldenmatch/docs/screenshots/web/web-inspector.png)](https://github.com/benseverndev-oss/goldenmatch/wiki/Web-UI)\n\n

_{Pair drilldown in the web workbench: cluster members, field-level diff, and a one-line NL explanation per pair. pip install goldenmatch[web] then goldenmatch serve-ui <project>. More screenshots →}

\n\n```bash\n# Headline package: dedupe a CSV in 30 seconds\npip install goldenmatch && goldenmatch dedupe customers.csv\n\n# TypeScript / Edge runtimes\nnpm install goldenmatch\n```\n\n\n> **🆕 v2.0.0** — **GoldenMatch 2.0.0: the first backwards-incompatible major.** It removes four deprecation-window items, each shipped with a 1.x runway: the legacy `:hash:` identity lookup bridge + `GOLDENMATCH_IDENTITY_ID_SCHEME` (run `goldenmatch identity migrate-ids` before upgrading; un-fingerprintable rows keep their `:hash:` id), the `GOLDENMATCH_CLUSTER_FRAMES_OUT` gate + legacy dict cluster path (`build_clusters` stays as a frames-backed adapter), and the `cheapest_healthy` / `_scale_aware_backend` shims. Pipeline behavior is output-equivalent. Migration guide: [Migrating to v2](https://docs.bensevern.dev/goldenmatch/migrating-to-v2).\n>\n> **v1.30.0** — **Zero-training Fellegi-Sunter now beats hand-rolled, expert-tuned Splink, head-to-head and reproducibly.** On one shared evaluator across every dataset Splink scores, GoldenMatch's probabilistic auto-config wins on all of them: `historical_50k` pairwise F1 **0.778 vs 0.757** (cluster-level B³ **0.844 vs 0.789**), `febrl3` **0.991 vs 0.965**, `synthetic_person` **0.998 vs 0.996** — made reproducible by an EM training-pair determinism fix (#829). Full bake-off: `docs/benchmarks/2026-06-09-splink-bakeoff.md`.\n>\n> **v1.26.0** — **100M records, distributed, on a 4-worker Ray cluster — verified.** The distributed Phase-5 pipeline (`GOLDENMATCH_DISTRIBUTED_PIPELINE=2`) now runs a full 100,000,000-row dedupe end to end in ~213 s with the driver process peaking at 0.30 GB RSS. The unlock was removing every driver-side collect from the pipeline (scoring -> per-partition local connected-components -> distributed join -> distributed golden build + write), so nothing funnels back to a single node.\n\n\n---\n\n## Why a suite?\n\nEach tool stands alone, but they compose into a single pipeline:\n\n```mermaid\nflowchart LR\n raw([raw rows])\n golden([golden records])\n\n subgraph orchestration [\"GoldenPipe orchestrates\"]\n direction LR\n infermap[InferMap]\n goldencheck[GoldenCheck]\n goldenflow[GoldenFlow]\n goldenmatch[GoldenMatch]\n infermap --> goldencheck --> goldenflow --> goldenmatch\n end\n\n raw --> infermap\n goldenmatch --> golden\n```\n\n| Step | Role |\n|---|---|\n| **InferMap** | schema mapping — auto-aligns columns across heterogeneous sources |\n| **GoldenCheck** | profile + validate — encoding, format, anomaly detection |\n| **GoldenFlow** | standardize + transform — phone, date, address, categorical normalization |\n| **GoldenMatch** | dedupe + cluster + survivorship — fuzzy / exact / probabilistic / LLM |\n| **GoldenAnalysis** | analysis + reporting — one exportable report over any stage's output, plus cross-run regression detection |\n| **GoldenPipe** | orchestrator — declarative YAML pipeline wiring the steps |\n\n- **Zero-config defaults that admit when they're unsure** — every step has a self-verifying preflight + postflight; results carry an inspectable report instead of failing silently.\n- **96.4% F1 on DBLP-ACM out of the box** for entity resolution — and the opt-in Fellegi-Sunter engine **beats hand-rolled, expert-tuned Splink** head-to-head on every dataset Splink scores (`historical_50k` pairwise F1 **0.778 vs 0.757**, cluster-level B³ **0.844 vs 0.789**; one shared evaluator, [reproducible bake-off](https://github.com/benseverndev-oss/goldenmatch/blob/main/docs/benchmarks/2026-06-09-splink-bakeoff.md)).\n- **Learning Memory** — corrections persist across runs and re-anchor across row reorders, so the system stops needing the same correction twice (GoldenMatch v1.6.0; off by default).\n- **Identity Graph** — a durable graph layer above run-local clusters: stable `entity_id`s that survive across runs, an append-only event log, and create / absorb / merge / split semantics, surfaced on the CLI, REST, MCP, and SQL interfaces (the Identity Graph v2 feature, shipped in GoldenMatch v1.15).\n- **Privacy-preserving record linkage** — match across organizations without sharing raw data (PPRL, 92.4% F1 on FEBRL4).\n- **AI-native by design** — every package ships an MCP server, a REST API, and an A2A agent surface. 50+ MCP tools across the suite, including `auto_configure` + `controller_telemetry` for v1.7-v1.12 introspection.\n- **AutoConfigController visible everywhere** (v1.7-v1.12 surface-parity arc) — web `ControllerPanel`, TUI `Ctrl+A`, CLI `goldenmatch autoconfig`, REST `/autoconfig` + `/controller/telemetry`, Postgres `goldenmatch_autoconfig` + `gm_telemetry`, DuckDB UDFs, MCP/A2A telemetry tools. One JSON shape across every interface.\n- **Polyglot parity** — the full suite ships on **npm** (goldenmatch, goldencheck, goldenflow, goldenanalysis, infermap, goldenpipe) alongside PyPI; the TypeScript and Python implementations track the same outputs to 4-decimal precision via a cross-language parity harness.\n- **Edge-safe, with optional native speed** — the TypeScript cores are dependency-free and `node:*`-free, so they run in browsers, Cloudflare Workers, Vercel Edge, and Deno. An **opt-in WebAssembly backend** (`await enableWasm()` / `enableAnalysisWasm()`) swaps in the *same* pyo3-free Rust kernels the Python wheels and the SQL UDFs use — pure-TS stays the default and the byte-identical fallback, so default users download zero wasm bytes.\n- **SQL-native, both engines at parity** — the same functions run inside **PostgreSQL** (pgrx extension) and **DuckDB**: dedupe / match / score / auto-config + telemetry / identity graph, plus data profiling, `evaluate`, Fellegi-Sunter probabilistic scoring, and GoldenFlow transforms.\n- **Production paths** — Postgres sync, daemon mode, lineage tracking, review queues, dbt integration, GitHub Actions, and a Rust extension layer for Postgres / DuckDB.\n\n---\n\n## The Suite\n\n| Package | Lang | What it does | Install |\n|---|---|---|---|\n| **[GoldenMatch](https://github.com/benseverndev-oss/goldenmatch/blob/main/packages/python/goldenmatch/README.md)** 🟡 | Python · TS | Zero-config entity resolution. Fuzzy + exact + probabilistic + LLM. Headline package. | `pip install goldenmatch` · `npm i goldenmatch` |\n| **[GoldenCheck](https://github.com/benseverndev-oss/goldenmatch/blob/main/packages/python/goldencheck/README.md)** | Python · TS | Data-quality scanning: encoding, Unicode, format validation, anomaly detection. | `pip install goldencheck` · `npm i goldencheck` |\n| **[GoldenFlow](https://github.com/benseverndev-oss/goldenmatch/blob/main/packages/python/goldenflow/README.md)** | Python · TS | Transforms & standardizers: phone, date, address, categorical normalization. | `pip install goldenflow` · `npm i goldenflow` |\n| **[GoldenPipe](https://github.com/benseverndev-oss/goldenmatch/blob/main/packages/python/goldenpipe/README.md)** | Python · TS | Orchestrator that wires Check → Flow → Match into one declarative pipeline. | `pip install goldenpipe` · `npm i goldenpipe` |\n| **[InferMap](https://github.com/benseverndev-oss/goldenmatch/blob/main/packages/python/infermap/README.md)** | Python · TS | Schema mapping engine — auto-aligns columns across heterogeneous sources. | `pip install infermap` · `npm i infermap` |\n| **[GoldenAnalysis](https://github.com/benseverndev-oss/goldenmatch/blob/main/packages/python/goldenanalysis/README.md)** | Python · TS | Cross-cutting analysis & reporting — consumes any stage's typed artifacts (or a raw DataFrame) and emits a unified, exportable `AnalysisReport`; optional Rust / WASM `histogram`+`quantile` kernels. | `pip install goldenanalysis` · `npm i goldenanalysis` |\n| **[goldenmatch-extensions](https://github.com/benseverndev-oss/goldenmatch/blob/main/packages/rust/extensions/README.md)** | Rust | Postgres extension (pgrx) + DuckDB UDFs. SQL-native fuzzy matching. | source build |\n| **[dbt-goldensuite](https://github.com/benseverndev-oss/goldenmatch/blob/main/packages/python/goldenmatch/dbt-goldensuite/README.md)** | dbt · Python | dbt package — quality-gate tests, correction CRUD macros + GoldenCheck assertions for warehouse models. | `pip install dbt-goldensuite` |\n| **[goldencheck-action](https://github.com/benseverndev-oss/goldenmatch/blob/main/packages/actions/goldencheck/README.md)** | YAML | GitHub Action — fail PRs that introduce data-quality regressions. | Marketplace |\n\n> Headline pitch and the deepest docs live in **[packages/python/goldenmatch/README.md](https://github.com/benseverndev-oss/goldenmatch/blob/main/packages/python/goldenmatch/README.md)** (~1,300 lines, full feature list, CLI, architecture, benchmarks).\n\n---\n\n## Choose your path\n\n| I want to... | Go here |\n|---|---|\n| Deduplicate a CSV right now | [`packages/python/goldenmatch`](https://github.com/benseverndev-oss/goldenmatch/blob/main/packages/python/goldenmatch/README.md#quick-start) |\n| Use from Claude Desktop / Code | [`packages/python/goldenmatch` — MCP](https://github.com/benseverndev-oss/goldenmatch/blob/main/packages/python/goldenmatch/README.md#remote-mcp-server) |\n| Edit rules in a browser, label pairs, compare runs | [`packages/python/goldenmatch` — Web UI](https://github.com/benseverndev-oss/goldenmatch/blob/main/packages/python/goldenmatch/README.md#web-ui) |\n| Build AI agents that deduplicate | [ER Agent / A2A wiki page](https://github.com/benseverndev-oss/goldenmatch/wiki/ER-Agent) |\n| Profile data quality before matching | [`packages/python/goldencheck`](https://github.com/benseverndev-oss/goldenmatch/blob/main/packages/python/goldencheck/README.md) |\n| Standardize messy fields (phone, date, address) | [`packages/python/goldenflow`](https://github.com/benseverndev-oss/goldenmatch/blob/main/packages/python/goldenflow/README.md) |\n| Run the full pipeline declaratively | [`packages/python/goldenpipe`](https://github.com/benseverndev-oss/goldenmatch/blob/main/packages/python/goldenpipe/README.md) |\n| Map columns across schemas | [`packages/python/infermap`](https://github.com/benseverndev-oss/goldenmatch/blob/main/packages/python/infermap/README.md) |\n| Analyze + report across stages and runs | [`packages/python/goldenanalysis`](https://github.com/benseverndev-oss/goldenmatch/blob/main/packages/python/goldenanalysis/README.md) |\n| Write TypeScript / Node.js / Edge (browser, Workers; optional WASM) | [`packages/typescript/goldenmatch`](https://github.com/benseverndev-oss/goldenmatch/blob/main/packages/typescript/goldenmatch/README.md) |\n| Match in Postgres / DuckDB SQL | [`packages/rust/extensions`](https://github.com/benseverndev-oss/goldenmatch/blob/main/packages/rust/extensions/README.md) |\n| Add data-quality gates to dbt | [`packages/python/goldenmatch/dbt-goldensuite`](https://github.com/benseverndev-oss/goldenmatch/blob/main/packages/python/goldenmatch/dbt-goldensuite/README.md) |\n| Block bad data in GitHub PRs | [`packages/actions/goldencheck`](https://github.com/benseverndev-oss/goldenmatch/blob/main/packages/actions/goldencheck/README.md) |\n| Run as Airflow DAGs | [`examples/airflow/`](https://github.com/benseverndev-oss/goldenmatch/blob/main/examples/airflow/README.md) — 12 drop-in DAGs |\n| Run from a single MCP container | [`docker run ghcr.io/benseverndev-oss/goldensuite-mcp:latest`](https://github.com/benseverndev-oss/goldenmatch/blob/main/packages/python/goldensuite-mcp/README.md) |\n| Pull every Suite container | [GitHub Packages](https://github.com/benzsevern?tab=packages) |\n\n---\n\n## Quick examples\n\n### Python — dedupe in 30 seconds\n\n```python\nimport goldenmatch as gm\n\n# Zero-config\nresult = gm.dedupe(\"customers.csv\")\nprint(result) # DedupeResult(records=5000, clusters=847, match_rate=12.0%)\nresult.golden.write_csv(\"deduped.csv\")\n\n# Or be explicit\nresult = gm.dedupe(\"customers.csv\",\n exact=[\"email\"],\n fuzzy={\"name\": 0.85, \"zip\": 0.95},\n blocking=[\"zip\"],\n threshold=0.85)\n```\n\n### TypeScript — edge-safe core\n\n```typescript\nimport { dedupe } from \"goldenmatch\";\n\nconst result = dedupe(rows, {\n fuzzy: { name: 0.85 },\n blocking: [\"zip\"],\n threshold: 0.85,\n});\nconsole.log(result.stats); // { totalRecords, totalClusters, matchRate, ... }\n```\n\nRuns in browsers, Vercel Edge, Cloudflare Workers, Deno — and optionally swaps in the Rust `score-core` kernel via `await enableWasm()`. ~940 tests, strict TypeScript (`noUncheckedIndexedAccess`, `exactOptionalPropertyTypes`).\n\n### Web workbench — browser UI for matching\n\n```bash\npip install 'goldenmatch[web]'\ngoldenmatch serve-ui my-project # opens http://localhost:5050\n```\n\n![GoldenMatch web UI](https://raw.githubusercontent.com/benseverndev-oss/goldenmatch/main/packages/python/goldenmatch/docs/screenshots/web/web-inspector.png)\n\nEdit rules with live validation, preview against a sampled slice, label pairs\n(mirrored into Learning Memory automatically), compare runs (CCMS), sweep\nparameters, browse the corrections store. Single-process localhost workbench\nshipped as the optional `[web]` extra.\n\n### Composed pipeline\n\n```python\nimport goldenpipe as gp\n\npipeline = gp.Pipeline.from_yaml(\"pipeline.yaml\") # check → flow → match\nresult = pipeline.run(\"customers.csv\")\nresult.report.write_html(\"report.html\")\n```\n\n**More**: [`examples/`](https://github.com/benseverndev-oss/goldenmatch/blob/main/examples/README.md) has runnable demos for every Suite scenario:\n[Python](https://github.com/benseverndev-oss/goldenmatch/blob/main/examples/python/README.md) (quickstart, full pipeline, customer 360, PPRL, review workflow, MCP client) ·\n[TypeScript](https://github.com/benseverndev-oss/goldenmatch/blob/main/examples/typescript/README.md) (quickstart, Vercel Edge route, MCP client) ·\n[Airflow DAGs](https://github.com/benseverndev-oss/goldenmatch/blob/main/examples/airflow/README.md) (12 production-shaped pipelines).\n\n---\n\n## Use cases (real-world pipelines)\n\nReproducible end-to-end pipelines running GoldenMatch on public data at scale, each with measured headline numbers vs baselines:\n\n- 🕵️ **[goldenmatch-shell-company-network](https://github.com/benseverndev-oss/goldenmatch-shell-company-network)** — investigative ER across ICIJ Offshore Leaks + OpenSanctions + GLEIF + UK PSC + UK disqualified-directors. Confidence-weighted graph, structure mining, named investigative candidates. **−62.5% analyst-hours to triage** vs single-source baselines; +133% adversarial perturbation recovery.\n- 🛡️ **[goldenmatch-vuln-attribution](https://github.com/benseverndev-oss/goldenmatch-vuln-attribution)** — cross-database ER on 6.1M OSS vulnerability records across 40 sources (OSV, GHSA, PyPA, RustSec, Go vulndb, EPSS, CISA KEV, CVE Project bulk). **6,126,895 records → 847,475 canonical vulns** in ~5 minutes end-to-end on a single 64GB runner via the full Golden Suite (Check + Flow + Match + Pipe).\n- ⚖️ **[goldenmatch-sanctions-reconciliation](https://github.com/benseverndev-oss/goldenmatch-sanctions-reconciliation)** — cross-list coverage analysis on 85 public sanctions lists across 50+ jurisdictions via OpenSanctions, plus 10-year OFAC SDN history and PEP/crypto cross-analysis. Coverage-gap benchmark for any sanctions-screening vendor.\n\n---\n\n## Install variants\n\nGoldenMatch ships fat optional extras so you only pay for what you use:\n\n```bash\npip install goldenmatch # core (CSV in, CSV out) + native acceleration on common platforms\npip install goldenmatch[native] # back-compat alias; native is already default on common platforms\npip install goldenmatch[embeddings] # + sentence-transformers, FAISS\npip install goldenmatch[llm] # + Claude / OpenAI for LLM boost\npip install goldenmatch[postgres] # + Postgres sync\npip install goldenmatch[snowflake] # + Snowflake connector\npip install goldenmatch[bigquery] # + BigQuery connector\npip install goldenmatch[databricks] # + Databricks connector\npip install goldenmatch[salesforce] # + Salesforce connector\npip install goldenmatch[duckdb] # + DuckDB out-of-core backend\npip install goldenmatch[ray] # + Ray distributed backend (50M+ rows)\npip install goldenmatch[quality] # + GoldenCheck integration\npip install goldenmatch[transform] # + GoldenFlow integration\npip install goldenmatch[mcp] # + MCP server for Claude Desktop\npip install goldenmatch[agent] # + A2A agent (aiohttp)\npip install goldenmatch[web] # + localhost browser workbench (FastAPI + React)\n\ngoldenmatch setup # interactive wizard: GPU, API keys, database\n```\n\nSister packages compose: `pip install goldenpipe[full]` brings in Check + Flow + Match together.\n\n---\n\n## Remote MCP Server\n\nGoldenMatch is hosted as an MCP server on [Smithery](https://smithery.ai/servers/benzsevern/goldenmatch) — connect from any MCP client without installing anything.\n\n```json\n{\n \"mcpServers\": {\n \"goldenmatch\": {\n \"url\": \"https://goldenmatch-mcp-production.up.railway.app/mcp/\"\n }\n }\n}\n```\n\n50+ MCP tools across the suite: deduplicate, match, explain, review, link privately, configure, scan quality, transform, synthesize golden records, and manage Learning Memory corrections.\n\n---\n\n## Container images\n\nEvery Suite package ships as a multi-arch container image (linux/amd64 + linux/arm64) on GitHub Container Registry. Pull anonymously, no auth needed:\n\n```bash\n# One container, every Suite tool — the convenience option\ndocker run -p 8300:8300 ghcr.io/benseverndev-oss/goldensuite-mcp:latest\n\n# Per-package containers — narrower deployments\ndocker run -p 8200:8200 ghcr.io/benseverndev-oss/goldenmatch-mcp:latest\ndocker run -p 8100:8100 ghcr.io/benseverndev-oss/goldencheck-mcp:latest\ndocker run -p 8150:8150 ghcr.io/benseverndev-oss/goldenflow-mcp:latest\ndocker run -p 8250:8250 ghcr.io/benseverndev-oss/goldenpipe-mcp:latest\ndocker run -p 8400:8400 ghcr.io/benseverndev-oss/infermap-mcp:latest\n\n# Postgres + extension preinstalled\ndocker run -e POSTGRES_PASSWORD=secret ghcr.io/benseverndev-oss/goldenmatch-extensions:latest\n```\n\nTags:\n- `:latest` — current `main`\n- `:main-` — every push to main, immutable\n- `:vX.Y.Z` and `:vX.Y` — pushed when a `-vX.Y.Z` tag is created\n\nSee [`packages/python/goldensuite-mcp/README.md`](https://github.com/benseverndev-oss/goldenmatch/blob/main/packages/python/goldensuite-mcp/README.md) for the aggregator's tool-collision behaviour.\n\n---\n\n## Airflow\n\n12 drop-in DAGs at [`examples/airflow/`](https://github.com/benseverndev-oss/goldenmatch/blob/main/examples/airflow/README.md), grouped by lifecycle stage:\n\n| Group | DAGs |\n|---|---|\n| **Core pipeline** | `daily_dedupe`, `incremental_match`, `warehouse_native` (Snowflake), `customer_360` (multi-source) |\n| **Privacy** | `pprl_linkage` (two-party PPRL) |\n| **Onboarding & monitoring** | `schema_align_and_load`, `schema_drift_alarm`, `quality_gate` |\n| **Feedback loop** | `review_worker`, `active_learning` |\n| **Operationalize** | `reverse_etl` (Salesforce/HubSpot), `backfill` |\n\nTaskFlow API, Airflow 2.7+ (compatible with 3.x). Each DAG has tunable knobs at the top, idempotent retries, and is marker-protected against double-processing. Drop the file you want into your Airflow `dags/` folder.\n\n---\n\n## Repository layout\n\n```\ngoldenmatch/\n├── packages/\n│ ├── python/\n│ │ ├── goldenmatch/ # entity resolution — headline package\n│ │ ├── goldencheck/ # data quality scanning\n│ │ ├── goldenflow/ # transforms & standardizers\n│ │ ├── goldenpipe/ # orchestrator\n│ │ ├── infermap/ # schema mapping\n│ │ └── goldenanalysis/ # cross-cutting analysis & reporting\n│ ├── typescript/\n│ │ ├── goldenmatch/ # full TS port (edge-safe core)\n│ │ ├── goldencheck/ # TS implementation\n│ │ ├── goldencheck-types/ # shared TS types\n│ │ ├── goldenflow/ # TS transforms\n│ │ ├── infermap/ # TS schema mapping\n│ │ └── goldenanalysis/ # TS analysis & reporting (edge-safe + WASM)\n│ ├── rust/\n│ │ └── extensions/ # Postgres pgrx + DuckDB UDFs (own Cargo workspace)\n│ ├── python/goldensuite-mcp/ # aggregator MCP server (one container, all tools)\n│ ├── dbt/goldencheck/ # dbt package\n│ └── actions/goldencheck/ # GitHub Action\n├── examples/\n│ ├── python/ # 6 runnable Python scripts (quickstart → MCP)\n│ ├── typescript/ # 3 TS scripts (quickstart, Vercel Edge, MCP)\n│ └── airflow/ # 12 drop-in Airflow DAGs\n├── docs/superpowers/ # design specs and implementation plans\n├── justfile # install / test / lint / build, all languages\n├── pyproject.toml # uv workspace (root)\n├── pnpm-workspace.yaml # TypeScript pnpm workspace (Turborepo)\n├── package.json # root scripts + pnpm workspace root\n└── .github/workflows/ci.yml\n```\n\n### Workspaces (Cargo vs pnpm)\n\n- **Cargo — no root workspace.** `packages/rust/extensions/` is itself a Cargo workspace (the `postgres` crate is excluded for pgrx-specific build requirements). Cargo doesn't allow nested workspaces sharing members, so Cargo commands run from inside `packages/rust/extensions/`.\n- **TypeScript — a single pnpm workspace.** `packages/typescript/*` form one pnpm + Turborepo workspace (see [TypeScript dev setup](#typescript-dev-setup-pnpm--turborepo)). `.npmrc` pins `node-linker=hoisted`, giving a flat `node_modules` that avoids the Windows symlink issues an earlier per-package layout hit.\n\n### Build / test / lint everything\n\n```bash\njust install # uv sync + per-package npm install + cargo fetch\njust test # all languages\njust lint\njust build\n```\n\n---\n\n## Reproducing benchmarks\n\nPublished GoldenMatch numbers (DQbench composite 91.04, DBLP-ACM 0.9641 F1, Febrl3 0.9443 F1, NCVR 0.9719 F1) map back to a single committed runner: `scripts/run_benchmarks.py`. See [`docs/reproducing-benchmarks.md`](https://github.com/benseverndev-oss/goldenmatch/blob/main/docs/reproducing-benchmarks.md) for per-number commands, dataset URLs, expected output (with tolerance), variance notes (deterministic vs LLM-augmented), and a copy-pasteable one-click reproduction snippet for the DQbench composite. The same runner powers the weekly `benchmarks.yml` workflow.\n\n## Scale envelope\n\n\"How big can this handle?\" is answered in [`docs/scale-envelope.md`](https://github.com/benseverndev-oss/goldenmatch/blob/main/docs/scale-envelope.md): per-backend ranges (Polars in-memory < 500K, DuckDB out-of-core 500K - 50M, Ray distributed >= 50M), block-size failure modes, candidate-pair math, and a single-page decision tree for picking a backend.\n\n**Verified at the top end:** a full **100,000,000-row** GoldenMatch dedupe on a 5-node Ray cluster (`e2-standard-16`, 80 CPU) in **9.2 min** (554 s), **20,000,000 golden records recovered exactly**, driver process peak **0.36 GB RSS** — the default distributed path is now **recall-complete** (blocking-key shuffle scoring + a distributed randomized-contraction WCC), so duplicates merge correctly *no matter how the input is partitioned*, and it stays driver-collect-free end to end (#844). A faster **per-partition path** is available via `GOLDENMATCH_DISTRIBUTED_BLOCK_SHUFFLE=0` (driver-collect-free, ~213 s on a 4-worker run) for inputs where duplicates already co-locate within partitions — but it under-merges when a cluster's members land in different input partitions, which is why recall-complete is the default. Recipe in [`packages/python/goldenmatch/configs/distributed-100m.yaml`](https://github.com/benseverndev-oss/goldenmatch/blob/main/packages/python/goldenmatch/configs/distributed-100m.yaml).\n\n---\n\n## Contributing\n\n- Feature work goes on `feature/` branches; merge via squash PR.\n- PR title format: `feat: `, `fix: `, `docs: `.\n- Tests must pass on all three languages where the change applies; the parity harness in `packages/typescript/goldenmatch/tests/parity/` enforces 4-decimal-tolerance Python ↔ TypeScript scorer parity.\n- See `docs/superpowers/specs/` for design rationale on architectural decisions.\n\n### TypeScript dev setup (pnpm + Turborepo)\n\nThe TypeScript packages live in a single pnpm workspace orchestrated by Turborepo. From the repo root:\n\n```bash\ncorepack enable # one-time, picks up pnpm@9.15.0 from package.json\npnpm install # installs all workspace packages\npnpm turbo run build test typecheck lint # full pipeline (cached after first run)\npnpm --filter goldenmatch test # single package\n```\n\n**Windows: enable Developer Mode for pnpm.** `pnpm install` creates symlinks under `node_modules/`. Settings → For Developers → Developer Mode → On. If you see `EPERM: operation not permitted, symlink ...` during install, Dev Mode is off.\n\nIf `corepack enable` fails (often needs an admin shell on Windows), the fallback is `npm i -g pnpm@9.15.0` — functionally equivalent.\n\n---\n\n## History\n\nThis repository was formed on **2026-05-01** by folding 8 sibling repos into the existing `goldenmatch` repo using `git filter-repo`. Full commit history is preserved for every source. See [`docs/superpowers/specs/2026-05-01-goldenmatch-monorepo-fold-in-design.md`](https://github.com/benseverndev-oss/goldenmatch/blob/main/docs/superpowers/specs/2026-05-01-goldenmatch-monorepo-fold-in-design.md) for the design rationale and [`docs/superpowers/plans/2026-05-01-goldenmatch-monorepo-fold-in.md`](https://github.com/benseverndev-oss/goldenmatch/blob/main/docs/superpowers/plans/2026-05-01-goldenmatch-monorepo-fold-in.md) for the step-by-step migration plan.\n\n---\n\n## Author & License\n\nBuilt by **[Ben Severn](https://bensevern.dev)**.\n\nMIT — see [LICENSE](https://github.com/benseverndev-oss/goldenmatch/blob/main/LICENSE).\n","changelogMarkdown","videoKey","requiredSecretsJson","publishedAt","priceCents","trialDays",0,"categoryDisplayName","Data & Analytics","upstream",{"_95":96,"_97":98,"_99":100,"_101":102,"_103":104},"requiredSecrets",[],"tags",[],"assets",[],"stats",{"_93":80,"_94":80},"ratingAvg","reviewCount","topics",[105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124],"license","MIT","language","Python","registryName","io.github.benseverndev-oss/goldenmatch","homepage","https://bensevern.dev/","active-learning","agent","airflow","auto-config","data-engineering","data-quality","deduplication","entity-resolution","fuzzy-matching","human-in-the-loop","llm","mcp-server","polars","pprl","privacy-preserving","python","record-linkage","splink","typescript","zero-config","ENV",{"_127":128,"_129":130,"_131":132},"ENVIRONMENT","production","TURNSTILE_SITE_KEY","0x4AAAAAAB8-rEI2gNDf_QhU","SENTRY_DSN","https://45fd1229f78089d5c3ee78be128030e1@o4510583261757440.ingest.de.sentry.io/4510584253448272"]