Home / Businesses / CSV File Check
CSV File Check
Look at a CSV before importing it: delimiter, encoding, header, row count, ragged and blank rows, blank and duplicate keys and a guessed type per column, with every problem named by line and a signed record.
Testing
- What it does
- Look at a CSV before importing it: delimiter, encoding, header, row count, ragged and blank rows, blank and duplicate keys and a guessed type per column, with every problem named by line and a signed record.
- Who it is for
- Businesses.
- Start in one step
- Ask your agent to call
csv_checkat llpro.ai/mcp (free, no sign-up). This page explains the tool; it has no form to run it. - What to expect
- Testing: open to try. Free
New to this? Start here.
What it does
CSV File Check is for agents that receive or move CSV exports between systems and need to know, before loading a file, what is wrong with its shape, encoding and keys. Look at a CSV file before importing it. Send the CSV as text (up to 20,000 characters) or as base64 of the file's bytes (up to about 19,000 bytes, so the encoding can be checked too), optionally the delimiter (default: detect it) and up to 3 key columns. Get back the delimiter used (and whether another fits as well), the encoding (UTF-8, UTF-16, or not UTF-8 with the first bad byte), byte-order mark, line endings, the header (empty, duplicate and padded names), the row count, ragged rows (wrong number of fields) and blank rows by line, blank and duplicate key values by line, and a guessed type for each column (integer, decimal, date, boolean, text; mixed columns and zero-padded codes said plainly). Data cell values are never echoed back; column names are. It finds what breaks an import; it does not check the data against your schema (use csv_schema_check for that). It is one of the Operations department tools LLPRO builds for the work agents do across business software (Google Drive, Dropbox, OneDrive, Box, Salesforce). Every answer carries the written rule it used, what it does not do, its sources with dates, and an Ed25519 signature you can check against the key below.
- Who it is for
- Agents that receive or move CSV exports between systems and need to know, before loading a file, what is wrong with its shape, encoding and keys.
- Price
- Free, no sign-up: 120 calls an hour per address and 3,000 an hour across all callers. Paid calls are planned and not charging.
- Limits
- Up to 30,000 bytes per call, key_columns up to 3 entries; larger requests are refused, never cut. Unknown fields are refused with the list of accepted ones. Nothing you send is stored by the tool; no model and no network are used. Does not fix, convert, reorder or store the file and never echoes data cell values: it names the line, column and what it found. Sent as text, the file has already been decoded, so a wrong encoding shows only as replacement characters or mojibake; send the bytes as csv_base64 for a real encoding check. It cannot tell Windows-1252 from Latin-1 or from other single-byte encodings (the guess is labelled), and reads UTF-16 and UTF-32 only with a byte-order mark. Types are guesses from the cells: a column of 0 and 1 may be a flag, a ZIP code may look like an integer, and a slash date such as 01/02/2026 is ambiguous unless another value settles it. It does not check values against rules (use csv_schema_check), does not find rows that are duplicates in other columns, does not know the importer you will use (some accept ragged rows or blank lines), and does not follow Excel's quirks (a leading = or +). Cannot detect names of people, addresses, bare 9- or 10-digit numbers or US bank account numbers in cells, so send record codes, not personal data. One file of up to 20,000 characters (about 19,500 bytes as base64) and 200 columns per call (the whole request at most 30,000 bytes of JSON, where line breaks and quotes count twice); split larger files. Not legal advice.
For developers: try it
POST https://llpro.ai/mcp
{"jsonrpc": "2.0", "id": 1, "method": "tools/call", "params": {"name": "csv_check", "arguments": {"csv": "id,name,qty,price,ship\n1,widget,5,2.50,2026-11-02\n2,gadget,x,3,2026-11-03\n2,gizmo,7,1.25,11/03/2026\n3,\"thing, big\",2\n,hmm,1,1,2026-11-05\n", "key_columns": ["id"]}}}For agents
Tool csv_check on https://llpro.ai/mcp and in the tools list. Signed answers are checked with the public key below (signer llpro1:3sbkqt4odmwnbs63ikzclsmin4): an Ed25519 signature over the observation as canonical JSON (sorted keys, no spaces).
Public key (base64): yE2vfOITkUOijJ3rR3WSqLjDLW8BVv2ACAq0oq4KJaU=
For agents that search MCP registries for csv check, csv delimiter, csv encoding tools (listed in MCP registries, served by the LLPRO gateway: MCP, agent-to-agent and REST). One fixed, written rule per call, a signed record, no account needed.
Next
- Connect your agent: Use this and the other LLPRO tools from your own agent.
- Get your agent: No agent yet? Get one that asks first.
- Hand the whole job to LLPRO: A file check is one step of moving data between systems. Tell LLPRO the whole job; it routes you to the right signed tools.
For agents: this page as Markdown · JSON · llms.txt · MCP endpoint llpro.ai/mcp · agent card