FineData

Install

Works with Claude Code, VS Code, Codex, Cursor, Windsurf, Claude Desktop. Read the repository before you give any tool access to your data.

npx agentmarket install connector/quality-network-finedata-mcp

About

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this server can do and what it asks for, from its tool list and install config. Checked Oct 25, 2026.

Tools

5 read · 2 write · 2 delete

Secrets

2 (API keys, tokens or passwords)

Install scripts

Not scanned yet

Runs as

Node.js package

Tools (9)

Read from the source code; the running server may differ. Descriptions are the server’s own words; hidden characters are shown as [U+…].

• batch_scrapeBatch Scrape

  Read up to 100 web pages in one batch (e.g. many product or listing URLs). `urls` may be a list of strings or objects {url, ...overrides}. Defaults formats=['markdown'] per item. Poll with get_batch_status.

• cancel_jobCancel Jobcan change or delete data

  Cancel a pending/processing async job.

• get_batch_statusGet Batch Statusread-only

  Get overall batch status and per-job results.

• get_job_statusGet Job Statusread-only

  Poll one async job by the job_id returned from scrape_async or batch_scrape. Not for finding pages or products. On completed, returns markdown (not raw HTML).

• get_usageGet Usageread-only

  Current period token usage (api_tokens_used). Not a request log. Does not include plan token limit — check the billing dashboard.

• list_jobsList Jobsread-only

  List this account's recent async scrape jobs (from scrape_async / batch_scrape). Not a history of synchronous scrape_url requests.

• scrape_asyncScrape Async

  Read a web page like scrape_url, but as an async job — for a slow page or heavy stealth. Submit an async scrape job (long-running / heavy stealth). Defaults formats=['markdown']. Poll with get_job_status. Always a GET request: POST, PUT, PATCH and DELETE go through send_http_request, which runs synchronously.

• scrape_urlScrape URLread-only

  Read one web page and return its content, by default as markdown: an article, a product page, a listing, a catalog page, or a site's own search-results URL. Always a GET request; the target is left unmodified. REQUIRED INPUT: call this tool with {"url":"https://..."}; put search terms in that URL's encoded query string (e.g. .../search?q=wireless%20headphones). It does not accept an argument named "query" or a bare natural-language search phrase — to look something up on a site, open that site's search URL. To send POST, PUT, PATCH or DELETE — submitting a form, calling a write API — use sen…

• send_http_requestSend HTTP Requestcan change or delete data

  Send a POST, PUT, PATCH or DELETE request through the same rendering and proxy pipeline, and return the response as markdown by default. Use this for submitting a form, calling a write API or deleting a resource. This request may change something at the other end, which is why it is a separate tool: to read a page, use scrape_url. Send the payload as a string in "body" and set the matching "headers", e.g. {"Content-Type": "application/json"}. Stealth modes consume more tokens than a plain request. Exact rates are in the documentation and via the get_usage tool.

Signals

Stars

0

Forks

0

Open issues

0

Weekly downloads

43

Last commit

Aug 14, 2026 (2 months ago)

Last release

Aug 14, 2026 (2 months ago)

Last package publish

Aug 14, 2026 (2 months ago)

From GitHub (fetched Oct 24, 2026) and the package registry.

Details

Registry name

io.github.quality-network/finedata-mcp

Version

0.3.1

Auth

Bearer token, API key

Packages

pypi finedata-mcpnpm @finedata/mcp-server

License

MIT

Repository

github.com/quality-network/finedata-mcp

Website

finedata.ai/mcp/

First seen

Oct 24, 2026

Last checked

Oct 24, 2026

Source

Official MCP Registrygithub-graphql