# GoldenMatch

> Entity resolution toolkit — deduplicate records and match across datasets using fuzzy, probabilistic, and LLM-powered scoring.

## Interfaces
- MCP Server: `goldenmatch mcp-serve` (10 tools: analyze, configure, deduplicate, match, explain, review, compare strategies, PPRL)
- Remote MCP: https://goldenmatch-mcp-production.up.railway.app/mcp/ (27 tools, Smithery: https://smithery.ai/servers/benzsevern/goldenmatch)
- A2A Server: `goldenmatch agent-serve --port 8200` (8 skills)
- CLI: `goldenmatch dedupe`, `goldenmatch match`, + 18 more commands
- Python API: `import goldenmatch` — `dedupe_df()`, `match_df()`, `score_strings()`, `evaluate()`, ~101 exports
- REST API: `goldenmatch serve` on port 8000

## Install
- `pip install goldenmatch`

## Docs
- [Architecture](CLAUDE.md): Full codebase guide with patterns, gotchas, and performance notes
- [PyPI](https://pypi.org/project/goldenmatch/): Package page
- [GitHub](https://github.com/benzsevern/goldenmatch): Source
- [Full API Reference](llms-full.txt): Extended documentation with all function signatures and usage patterns
