// MCP TOOL REFERENCE

GOLDENMATCH TOOLS

32 tools from the GoldenMatch MCP Server, categorised by risk level.

// ALL 32 GOLDENMATCH TOOLS

READ 24 tools

Read dedupe Deduplicate records in a file. Returns cluster counts and optional output path. Read explain_pair Explain why two records match (or don Read find_exact_matches Find exact matches on a field in a file. Returns pairs. Read find_fuzzy_matches Find fuzzy matches in a block of rows. Returns scored pairs. Read goldenmatch-js Entity resolution agent -- dedupe, match, profile, score, explain, evaluate. Read identity_conflicts List evidence edges marked Read identity_history Return the temporal event log for an identity. Read identity_list List identities, optionally filtered by dataset/status. Read identity_split Split a subset of records off an identity into a brand-new identity. The original keeps the remaining records. Read learn_thresholds Force a MemoryLearner pass over accumulated corrections. Returns the list of LearnedAdjustments produced (m... Read list_blocking_strategies List all blocking strategy names. Read list_corrections List stored Learning Memory corrections, optionally filtered by dataset. Returns id_a, id_b, decision, sour... Read list_scorers List all available similarity scorers. Read list_strategies List all golden-record survivorship strategies. Read list_transforms List all available field transforms. Read match Match a target file against a reference file. Returns matched/unmatched counts. Read memory_stats Return Learning Memory status: total correction count, last learn time, and current learned adjustments. Ch... Read profile Profile a dataset: per-column null rate, cardinality, inferred type, samples. Read read_file Read a CSV/JSON file and return the first N records. Read score Score similarity between two strings. Read score_pair Score two record objects across weighted fields. Returns a combined score. Read score_strings Score similarity between two strings using the requested scorer. Read server_info Return metadata about this GoldenMatch MCP server. Read suggest_config Suggest a shorthand dedupe config based on a profile of the dataset.

WRITE 5 tools

Write add_correction Add a pair correction to Learning Memory. Source is set to Write identity_merge Manually merge two identities. All records from Write identity_resolve Resolve a record_id to its durable identity. Returns the full identity view (members, evidence edges, recen... Write memory_export Return all corrections as a list of dicts (CSV-shaped). Caller is responsible for writing the file. Optiona... Write write_csv Write a list of record objects to a CSV file.

EXECUTE 3 tools

Execute build_clusters Group records into clusters given a file and matchkey definition. Execute evaluate Evaluate predicted pairs from a dedupe run against ground truth pairs. Execute explain_cluster Run dedupe on a file and explain the cluster containing the given row id.

// RUNNING THIS SERVER

The managed route: connect GoldenMatch through the PolicyLayer gateway — every tool call above is checked against your policy before it runs, with a full audit log.

DIRECT INSTALL (UNMANAGED) npx -y goldenmatch

// FAQ

How many tools does the GoldenMatch MCP server have? +

The GoldenMatch MCP server exposes 32 tools across 3 categories: Read, Write, Execute.

How do I enforce policies on GoldenMatch tools? +

Route the GoldenMatch server through the PolicyLayer gateway. Define allow, deny, or approval rules per tool in the dashboard — they are enforced on every call before it reaches the server.

What risk categories do GoldenMatch tools fall into? +

GoldenMatch tools are categorised as Read (24), Write (5), Execute (3). Each category has a recommended default policy.

Let agents act without letting them run wild.

Route your MCP servers through PolicyLayer and every tool call is checked against your policy before it runs — allow, deny, or require approval. Per-identity grants. Full audit log. Live in minutes.

SECURE YOUR MCP →

Free to start. No card required.

4,600+ MCP servers and 31,000+ tools scanned and risk-classified.