CSV (RFC 4180)

Comma-separated values — the format everyone thinks is simple.

16 capabilities 5 sections reference specification

Build against it

# give your agent the catalog — no account, no install
claude mcp add provespec -- npx -y provespec-mcp

# then, from any agent session
spec_get(slug: "csv")                → the capability tree
spec_checklist(slug: "csv")          → the questions to answer
spec_grade(slug: "csv", grades: [...])  → parity % + the gap list

# or just fetch it
curl https://provespec.com/catalog/csv/spec.json

Grade every capability below as yes · partial · no · na, with a file reference as evidence. Whatever you leave ungraded counts as missing — the gap list is the work queue.

Or just paste this to your agent

Use the ProveSpec MCP server. Fetch the "csv" spec with
spec_get, then walk spec_checklist and grade this repository against it —
one verdict per capability, each with a file reference as the note.
Call spec_grade with the results and give me the gap list as the plan.

If the server is not registered, add it first:
  claude mcp add provespec -- npx -y provespec-mcp

The specification

Grammar 6

✓ Records separated by CRLF the last record may omit it
✓ Fields separated by commas
✓ Optional header line same field count as records
✓ Fields may be double-quoted
✓ Quoted fields may contain commas, CRLF and quotes quotes doubled inside
✓ Spaces are part of the field not trimmed

Reading 5

✓ Configurable delimiter comma, tab, semicolon
✓ Configurable quote and escape characters
✓ Ragged rows reported, not silently padded
✓ BOM detected and stripped
✓ Streaming parse of large files constant memory

Writing 3

✓ Quote only when necessary minimal output
✓ Round-trips: parse then write reproduces the values
✓ Newline convention selectable CRLF or LF

Encoding 1

✓ UTF-8 assumed, other encodings declarable

Interop 1

✓ Excel dialect quirks documented leading =, long numbers, locale delimiters

Download spec.json All 53 specs