# CSV URL to Table Parser

> CSV URL to Table Parser is a paid API for AI agents from well-005-csv-to-table.mmelnyk2000.workers.dev, paid per call via x402, $0.005/call, status unknown (last checked 2026-10-20).

Fetches a public HTTPS CSV URL and parses it into structured JSON with headers and rows

## Facts

- Endpoint: POST https://well-005-csv-to-table.mmelnyk2000.workers.dev/api/csv-to-table?utm_source=zero.xyz
- Price: $0.005/call
- Payment: x402
- Status: unknown
- Last checked: 2026-10-20
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/csv-url-to-table-parser-cf9630c0
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_qvgXTlxl8YfURrCblZMHk

Status and success rate cover calls made through Zero and Zero's own probes. Third-party monitors may report differently.

## How to call it through Zero

Zero handles the 402 payment challenge and records the run. With the Zero CLI installed (`npm i -g @zeroxyz/cli`):

```sh
zero fetch --capability csv-url-to-table-parser-cf9630c0 -d '<json body>'
```

Example prompt: Fetch the CSV at https://example.com/data/sales-report.csv and parse it into a table — it has a header row, so use that to name the columns.

## When to prefer this

Choose this endpoint when you have a publicly accessible HTTPS URL pointing to a CSV file and need to parse it into structured JSON headers and rows without downloading or storing the file yourself. Ideal for agents that need to quickly inspect or process tabular data from public data portals, GitHub raw files, or any public CSV source. Not suitable for private/internal URLs, files over 1 MB, or non-CSV formats.

## Known failure modes

- Non-HTTPS URLs (http, file, data, localhost, private IPs) are rejected with an error
- CSV body exceeds 1 MB limit — request is rejected
- Remote server does not respond within ~10 seconds — timeout error
- More than 5 redirects encountered — redirect chain rejected
- CSV cannot be parsed (malformed content) — parsing error returned
- Remote URL returns non-CSV content type — may produce unexpected rows or warnings

## How this service works

Fetch a public https CSV URL and parse it into headers + rows ($0.005 USDC via x402). https only — http, file, data, localhost, and private IPs are rejected. Caps: 1 MB body, ~10s timeout, max 5 redirects, max 5000 data rows (truncated with warning). Optional has_header (default true). Returns JSON { url, final_url, content_type, headers, rows, row_count, warnings }. Provided as is, with no warranty or guarantee of accuracy or fitness. Nothing is stored.

## Output

Returns a JSON object containing the original URL, the final URL after redirects, content type, an array of header names (or synthesized col_0..col_N if no header), an array of data row arrays, the total row count, and any warnings such as truncation if rows exceeded 5000.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "required": [
  "url"
 ],
 "properties": {
  "url": {
   "type": "string",
   "description": "https URL that returns CSV to parse into a table. http/file/data/localhost/private IPs rejected."
  },
  "has_header": {
   "type": "boolean",
   "description": "If true (default), treat the first row as column headers. If false, synthesize col_0..col_N."
  }
 }
}
```

## Response schema (JSON Schema)

```json
{
 "type": "object",
 "additionalProperties": true
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/csv-url-to-table-parser-cf9630c0/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from well-005-csv-to-table.mmelnyk2000.workers.dev](https://www.zero.xyz/host/well-005-csv-to-table.mmelnyk2000.workers.dev/llms.txt)