InferMap
benzsevern/infermap
This connects Claude to an inference-driven schema mapper that figures out how messy CSV columns correspond to a clean target schema. You get seven scorers out of the box: exact match, alias dictionary, initialism detection, pattern types (email, phone, UUID), statistical profiling, fuzzy string similarity, and an optional LLM scorer. It runs the Hungarian algorithm to find optimal one-to-one assignments, then returns mappings with confidence scores and human-readable reasoning. The Python and TypeScript implementations share a golden-test parity suite, so mapping decisions are verified bit-for-bit across both. Reach for this when you're onboarding external data, normalizing CRM exports, or building ETL pipelines where column names are inconsistent and you need reproducible, explainable field matching without manual config.
- tools
- 4
- transport
- STDIO, HTTP
- last updated
- May 11, 2026
Install
From registry metadata$ npx add-mcp https://infermap-mcp-production.up.railway.app/mcp/ --name infermapRun in your terminal. add-mcp asks which agents to add it to (Claude Code, Codex, Cursor and more). Add -g to install it for every project.
Review the command, arguments and environment values before installing. MCP servers run with your local permissions.
Other ways
Claude Code
$ claude mcp add --transport http infermap https://infermap-mcp-production.up.railway.app/mcp/Run in your terminal. Add --scope user to make it available in every project.
Tools
Verified live · 4 toolsVerified live against the running server on Jun 10, 2026.
mapMap source columns to target schema using a weighted scorer pipeline with optimal 1:1 assignment. Returns mappings with confidence scores and human-readable reasoning.6 params
Map source columns to target schema using a weighted scorer pipeline with optimal 1:1 assignment. Returns mappings with confidence scores and human-readable reasoning.
Parameters* required
domainsarrayDomain dictionaries to load (e.g. ['healthcare', 'finance'])
min_confidencenumberMinimum confidence threshold (default 0.2)default: 0.2
schema_filestringPath to schema definition YAML file (optional)
source*stringPath to source data (CSV, Parquet, Excel, DB URI, schema YAML)
tablestringTable name for DB sources (optional)
target*stringPath to target data (same variety of inputs)
inspectInspect a data source — show fields, types, sample values, null rates, unique rates, and statistics.2 params
Inspect a data source — show fields, types, sample values, null rates, unique rates, and statistics.
Parameters* required
source*stringPath to data source (CSV, Parquet, Excel, DB URI, schema YAML)
tablestringTable name for DB sources (optional)
validateValidate that a source file's columns satisfy a saved mapping config. Reports missing source columns and unmapped required fields.3 params
Validate that a source file's columns satisfy a saved mapping config. Reports missing source columns and unmapped required fields.
Parameters* required
config*stringPath to mapping config YAML file
required_fieldsarrayTarget field names that must be mapped
source*stringPath to source data file
applyApply a saved mapping config to a source CSV, renaming columns according to the mapping and writing the result to an output file.3 params
Apply a saved mapping config to a source CSV, renaming columns according to the mapping and writing the result to an output file.
Parameters* required
config*stringPath to mapping config YAML file
output*stringOutput CSV file path
source*stringPath to source CSV file
Sponsors
Promote your product hereReach 125k+ visitors monthlyPromote your product hereReach 125k+ visitors monthly
Agent CaffeineSponsoredClose your lid, come back to finished work
SnugSponsoredThe clipboard manager for the AI eraPromote your product hereReach 125k+ visitors monthly
WaxSponsoredBlock anything with supervision, without erasing your iPhonePromote your product hereReach 125k+ visitors monthly
knobsSponsoredSee your site on any phone, any day, in any language
Agent CaffeineSponsoredClose your lid, come back to finished workSnugSponsoredThe clipboard manager for the AI eraPromote your product hereReach 125k+ visitors monthlyWaxSponsoredBlock anything with supervision, without erasing your iPhonePromote your product hereReach 125k+ visitors monthlyknobsSponsoredSee your site on any phone, any day, in any languageReadme
Moved. This repo has moved into the
benzsevern/goldenmatchmonorepo atpackages/python/infermap (and packages/typescript/infermap)/. This repo is archived; new development happens in the monorepo.
infermap
Inference-driven schema mapping engine.
Map messy source columns to a known target schema — accurately, explainably, with zero config.
Built by Ben Severn.
Details
- Repository
- github.com/benzsevern/infermap
- Website
- benzsevern.github.io/infermap
- Package
infermap- Transport
- STDIO, HTTP
- Registry
- MCP registry, active
- Tools verified
- Jun 10, 2026