CSV to JSON Schema
Read a CSV and emit a JSON Schema that describes each row as an object. Per-column types are inferred from observed values, with min/max for numerics and format detection (date, date-time, email, uri, uuid) for strings. Pair with json-schema-validate to lock incoming CSVs to a known shape.
About CSV to JSON Schema
CSV to JSON Schema reads a spreadsheet export and writes out a JSON Schema that treats every row as an object, inferring each column's type from the values it actually contains. Reach for it when you want to lock an incoming feed to a known shape instead of trusting whatever a partner uploads. It runs entirely in your browser, so the CSV never leaves your machine.
- Category
- convert
- Input
- Accepts: text/csv or text/plain.
- Output
- Outputs: application/json.
- Cost
- Free, runs in your browser
- Memory
- low
Common uses
- Generating a contract for a vendor CSV feed so downstream imports can reject malformed rows before they hit your database
- Inferring numeric min/max bounds from a sample export to catch out-of-range values in future uploads
- Auto-detecting that an 'id' column is really UUIDs and an 'email' column is really emails, then enforcing those string formats
- Bootstrapping a JSON Schema for an API endpoint that accepts CSV-derived records, without hand-writing field types
- Documenting the structure of a legacy data dump for teammates who have never seen the file
- Producing a schema to feed straight into a validator in a data-cleaning pipeline
Frequently asked questions
What input formats does it accept?
CSV files (text/csv) and plain text containing CSV data. The first row is treated as the header that names each column.
How are column types decided?
Types are inferred from the observed values across rows: numerics get min/max bounds, and strings are checked for date, date-time, email, uri, and uuid formats.
Does my CSV get uploaded anywhere?
No. Parsing and schema inference happen locally in your browser, so the data never touches a server.
Is there a row or file-size limit?
There is no hard cap; the practical limit is your browser's memory. Very large CSVs (hundreds of MB) may run slowly since the whole file is processed in-page.
What if a column has mixed or empty values?
Inference is based on the values it sees, so sparse or mixed columns produce a looser type. Feed a representative sample for the tightest schema.
Keywords
- csv
- json-schema
- jsonschema
- infer
- derive
- validate
- data