
Single-file, zero-dependency MCP server giving Claude, Cursor and any MCP client live web access.
Single-file, zero-dependency MCP server giving Claude, Cursor and any MCP client live web access.
·
Connect QuanticData to Claude, Cursor, and any MCP client. Gives an AI agent live web access — scrape, search, map, and crawl — through residential proxies with real-browser TLS fingerprints, so pages that block ordinary bots come back clean. It also hands the agent raw proxy endpoints of every type (residential, mobile, datacenter, ISP, IPv6) from your active plans, ready to plug into any HTTP client.
It calls the public QuanticData Scraper API with your own qd_live_ key,
so there are no internal secrets and you run it locally.
| Tool | What it does |
|---|---|
scrape | Scrape one URL → Markdown/HTML/text, including PDF/Office documents. Supports multi-format output, absolute link collection, JSON-LD metadata, structured CSS extraction, AI prompt/JSON-schema extraction, and mode: summary. |
unlock | Web Unlocker: replay any HTTP request (method, headers, body) through a residential exit with a real browser TLS fingerprint, retry on a fresh IP, escalate a blocked GET to a real browser. Returns the raw response (status, headers, body); a still-blocked page comes back flagged (blocked, blockClass, vendor), never as a silent 200. Interactive captchas are not solved (blockClass: "captcha"). |
seo_audit | Fetch a URL as a no-JS bot and fully rendered, return both SEO views + the diff (JS-only content, changed title/description, missing canonical) and bot-facing meta (robots, OG, JSON-LD). |
ai_visibility | Can ChatGPT, Claude, Perplexity, Google AI Overview and Bing Copilot read and cite a page? On-page audit (24 AI crawlers in robots.txt, Content-Signal, a fetch as GPTBot, noindex/nosnippet/noai, text without JS, JSON-LD entities, answer-shaped copy, dates with age, author, sources) scored per pillar with blockers, evidence and fixes — plus an optional citation panel that asks the engines and reports cited / mentioned / rank, share of voice and who wins the questions you are absent from. |
search | Structured Google/Bing/DuckDuckGo results. Set render: true for Google AI Overview, PAA, Knowledge Graph and other JS enrichments. |
search_and_read | SERP → fetch top pages → numbered citation-ready sources and one token-bounded context string ready for an AI prompt. |
search_bulk / search_bulk_status | Async multi-page pagination with merged organic results and page-one AI/zero-click enrichments. |
map | Fast URL discovery (sitemaps + homepage links), no full crawl. Compact by default: up to limit URLs (100) plus site-wide total and per-section summary; group_by: path for the path tree. |
crawl / crawl_status | Async BFS site crawl → Markdown per page; poll for progress. |
batch / batch_status | Scrape many URLs asynchronously; mode: summary for metadata-only items. Incremental polling via since cursor; page content only with include_content. |
create_dataset / dataset_status | Prompt-driven structured dataset collection with budget and row limits. |
list_collectors / run_collector / collector_run_status | Ready-made Collectors: run a versioned scraper with a semantic input (keyword + location, place id, product id, domain…) instead of URLs — Google Maps places, place reviews, Google Jobs/News/Shopping, product offers, hotels, local business leads, site contacts, company profile. Priced per delivered row; async runs poll by run_id, rows exportable as CSV. |
list_proxies | List your proxy services of every type — Residential Basic/Premium/Private, Mobile, Mobile V2, Datacenter, ISP, IPv6 — with bandwidth left, expiry and the orderId used to generate. |
generate_proxies | Ready-to-use proxy strings (credentials included) from any active plan: geo targeting (country/state/city/ISP/ASN), rotating or sticky sessions, HTTP or SOCKS5, several output formats. |
proxy_locations | Valid geo-targeting values per plan type: countries, states, cities, ASNs, or the full location tree with ISP codes. |
whitelist_ip | Manage IP-auth whitelisting (add/list/remove) for plans that support it, including the Mobile V2 IP-auth proxy list. |
report | Send feedback to the team from inside the agent: a bug (wrong/empty result, error, blocked page), a missing feature or collector, a question. Works without an API key. |
No install needed — npx fetches quanticdata-mcp on demand (Node.js 18+).
Claude Code (one command):
claude mcp add quanticdata \
-e QUANTICDATA_API_KEY=qd_live_your_key_here \
-- npx -y quanticdata-mcpClaude Desktop / Cursor / any MCP client (claude_desktop_config.json, .cursor/mcp.json, or .mcp.json):
{
"mcpServers": {
"quanticdata": {
"command": "npx",
"args": ["-y", "quanticdata-mcp"],
"env": { "QUANTICDATA_API_KEY": "qd_live_your_key_here" }
}
}
}The same server also runs as a hosted Streamable HTTP endpoint, for clients that prefer a URL over a local package:
https://api.quanticdata.io/mcpThe endpoint supports the MCP authorization spec (OAuth 2.1). In a client that supports it, add the URL above as a remote server and nothing else:
MCP · <client> appears on the
dashboard API keys page; delete it to disconnect the client.Under the hood: authorization server metadata at
https://app.quanticdata.io/.well-known/oauth-authorization-server,
protected resource metadata at
https://api.quanticdata.io/.well-known/oauth-protected-resource/mcp,
dynamic client registration (RFC 7591) and Client ID Metadata Documents,
PKCE (S256) required, 1-hour access tokens with rotating refresh tokens, tokens
bound to this endpoint (RFC 8707). Usage is billed to your balance like any key.
Clients that take a bearer token can skip OAuth: the key travels per request in
the Authorization header (or X-Api-Key), nothing is stored server-side.
curl -X POST https://api.quanticdata.io/mcp \
-H "Authorization: Bearer qd_live_your_key_here" \
-H "Content-Type: application/json" \
-H "Accept: application/json, text/event-stream" \
-d '{"jsonrpc":"2.0","id":1,"method":"tools/list"}'A request with neither a token nor a key gets 401 with a
WWW-Authenticate header pointing at the metadata above: that is what makes
OAuth-capable clients start the sign-in.
Self-hosting the endpoint is a second binary in this same package. With
QUANTICDATA_API_KEY set it serves that one account to callers that send no
credential:
QUANTICDATA_API_KEY=qd_live_… PORT=9310 npx -y quanticdata-mcp-remotenpm install
npm run buildThen point the client at the local build instead of npx:
{
"mcpServers": {
"quanticdata": {
"command": "node",
"args": ["/absolute/path/to/scraper-mcp/dist/index.js"],
"env": {
"QUANTICDATA_API_KEY": "qd_live_your_key_here"
}
}
}
}| Var | Default | Notes |
|---|---|---|
QUANTICDATA_API_KEY | — | Required. Your qd_live_ key. |
QUANTICDATA_API_BASE | https://api.quanticdata.io/v1 | Override for staging/self-host. |
QUANTICDATA_TOOLS | all 28 | Load only some tools: names (search,scrape) or a preset (lite, web, research, collectors, proxies). See below. |
Self-hosted remote endpoint only (quanticdata-mcp-remote), optional:
| Var | Default | Notes |
|---|---|---|
MCP_OAUTH_RESOURCE | https://api.quanticdata.io/mcp | This endpoint as an OAuth resource. |
MCP_OAUTH_ISSUER | https://app.quanticdata.io | Authorization server advertised in the resource metadata. |
MCP_OAUTH_INTROSPECT_URL | http://127.0.0.1:3090/api/v1/internal/oauth/introspect | Where OAuth access tokens are checked. |
MCP_INTERNAL_TOKEN | — | Secret for the introspection call. Unset = only API keys are accepted. |
The 28 tool definitions are about 14,000 tokens. A local model on Ollama's default 4K context cannot hold them, and a 32K context loses almost half its room before the first page. Load fewer tools:
| Preset | Tools | Definition tokens |
|---|---|---|
lite | search_and_read | ~360 |
web | search, search_and_read, scrape | ~4,200 |
research | web + map, batch, batch_status | ~5,100 |
collectors | list_collectors, run_collector, collector_run_status | ~830 |
proxies | list_proxies, generate_proxies, proxy_locations, whitelist_ip | ~1,800 |
Over stdio set QUANTICDATA_TOOLS=web. On the hosted endpoint add ?tools=web to the
URL and send your API key as a Bearer header:
https://api.quanticdata.io/mcp?tools=web. Names can be combined: ?tools=lite,collectors.
Full policy: https://quanticdata.io/privacy
What this server sends. It runs on your machine and talks only to the public
QuanticData API at https://api.quanticdata.io/v1, authenticated with
your own qd_live_ key. Each call carries the arguments you (or your agent)
passed — the target URL or query, and any extraction prompt. Nothing else on your
machine is read or transmitted: the server has no filesystem, shell or clipboard
access.
What we collect. Account data you give us (name, email, billing details) and a request log kept for billing, abuse prevention and support: target URL or query, timestamp, response status, bytes transferred and the API key used. We do not retain scraped page content beyond what is needed to return your result.
Data you collect through the service. You decide what public web data to collect and you are the controller of that data — we process it on your behalf only to fulfil your request. You are responsible for using the service lawfully, for respecting the terms of the sites you access, and for any personal data you collect through it.
Sharing. We do not sell personal data. We share it only with the processors that run the service — payment providers, email delivery, hosting, CDN and analytics — under contracts that bind them to protect it, and with the upstream proxy networks that route your traffic (connection metadata only, never your account details). We may disclose data where required by law.
Retention. Account data for as long as the account exists; request logs on a rolling window for billing and abuse investigation. You can request deletion at any time.
Your key. Claude Desktop stores it in the OS keychain and passes it to this server as an environment variable. It is never written to the bundle and never sent anywhere except the QuanticData API. Revoke or rotate it at https://app.quanticdata.io/api-keys.
Contact. [email protected]
Pick your client and paste the snippet. Each one is the same server, written the way that client expects it.
claude mcp add quanticdata -- npx -y quanticdata-mcp{
"mcpServers": {
"quanticdata": {
"env": {
"QUANTICDATA_API_KEY": ""
},
"args": [
"-y",
"quanticdata-mcp"
],
"command": "npx"
}
}
}code --add-mcp '{"name":"quanticdata","env":{"QUANTICDATA_API_KEY":""},"args":["-y","quanticdata-mcp"],"command":"npx"}'[mcp_servers.quanticdata]
command = "npx"
args = ["-y", "quanticdata-mcp"]Runs locally on your device. Your client starts the server itself, so nothing has to be hosted.
Paste this prompt into your agent. It reads this page and does the setup for you.
Read https://aiagentslisting.com/mcp/quanticdata-mcp-server to learn what the "QuanticData MCP Server" MCP server does and how to install it. Add it to my coding agent's MCP configuration as documented on that page, then confirm the server connects and list the tools it exposes.Agents can also browse this directory over MCP at https://aiagentslisting.com/api/mcp
claude mcp add quanticdata| Name | What it does |
|---|---|
scrape | Use this when the user wants the content of one web page. Fetches the page from a residential IP and returns it as clean Markdown (or HTML/text). It tries a plain HTTP request with a Chrome-compatible |
unlock | Use this when the user needs a site's raw HTTP response (a JSON API, a form POST, a page exactly as served) rather than scrape's Markdown. Sends the request (method, headers, body) from a residential |
generate_parser | Look at a page ONCE with an LLM and get back CSS selectors that extract the fields you asked for. Pass the returned `parser` as the `extract` argument on every later scrape of that same layout and no |
save_parser_preset | Store a generated parser under a name so it can be reused by id. Scrape later with scrape's `preset_id` instead of repeating the selectors, and every run is scored per field — when the recent success |
list_parser_presets | List your stored parser presets with their version, health stats and changelog. |
parser_preset_stats | How well a stored parser is still working: success rate per field, mean coverage over the recent runs, and whether it now counts as decayed (i.e. the site probably changed). |
heal_parser_preset | Regenerate a preset's selectors now (the manual trigger for the automatic repair). Refetches the source page and adopts new selectors ONLY if they extract more than the current ones — a heal that find |
seo_audit | Audit a URL's SEO in one call: fetches it twice — as a pure HTTP bot (no JS) and fully rendered — and returns both views (title, description, canonical, h1, word count) plus the diff (JS-only content, |
ai_visibility | Use this when the user asks whether AI assistants (ChatGPT, Claude, Perplexity, Google AI Overview, Bing Copilot) can find, read and cite a page; use seo_audit for classic Google indexing questions. O |
search | Run structured Google, Bing or DuckDuckGo searches through a residential proxy. Bing supports web, shopping, images, news, videos, places/maps and autocomplete over HTTP, including Copilot AI answers |
search_and_read | Search the live web, fetch the top organic pages as clean Markdown, and return citation-ready numbered sources plus one token-bounded `context` string ready for an AI prompt. Use this when the goal is |
map | Discover a site's URLs fast (robots.txt sitemaps + /sitemap.xml + homepage links) without a full crawl. Returns up to `limit` URLs (default 100) plus the site-wide `total` and a per-section `summary` |
crawl | Start an asynchronous BFS crawl of a site from a seed URL, converting each page to Markdown. Returns a job id — poll with crawl_status. |
crawl_status | Poll a crawl job for progress and the pages crawled so far. Polls are incremental: pass the previous response's `nextCursor` as `since` to receive only the pages crawled since your last poll. Pages om |
batch | Scrape many URLs asynchronously with shared options. Returns a job id — poll with batch_status. For SEO/status audits over many pages set mode 'summary': items carry metadata only (title, description, |
batch_status | Poll a batch job for progress and per-URL results. Polls are incremental: pass the previous response's `nextCursor` as `since` to receive only the items completed after your last poll. Items omit page |
search_bulk | Paginate ONE search query asynchronously and merge deduplicated organic results. Page-one AI Overview/PAA/Knowledge Graph/answer enrichments are retained; set render:true to request those Google JS bl |
search_bulk_status | Poll a bulk search job for progress and merged organic results. Polls are incremental: pass the previous response's `nextCursor` as `since` to receive only the organic results gathered after your last |
create_dataset | Build a structured dataset from a plain-language prompt. Quantic AI plans the search queries, searches Google/Bing/DuckDuckGo, maps the sites it finds and scrapes them into validated rows (CSV/JSON). |
dataset_status | Poll a dataset job for progress, the collection trace (steps) and the rows so far. Polls are incremental: pass the previous response's `nextCursor` as `since` to receive only rows delivered after your |
list_proxies | List the account's proxy services of every type — Residential Basic/Premium/Private, Mobile, Mobile V2, Datacenter (static or traffic-based), ISP, IPv6 — with plan type, bandwidth left, expiry, whitel |
generate_proxies | Generate ready-to-use proxy endpoint strings (credentials included) from one of the account's active proxy services — any type: residential, mobile, datacenter, ISP, IPv6. Supports geo targeting (coun |
proxy_locations | Discover valid geo-targeting values for a proxy plan type before calling generate_proxies: countries, states, cities, ASNs, or the full location tree (countries → regions → cities → ISPs). Use level ' |
whitelist_ip | Manage IP-auth whitelisting on a proxy service (Residential Basic, Datacenter, ISP, IPv6, Mobile): add or remove an IP, or list the current entries. A whitelisted machine uses the proxies without user |
list_collectors | Use this to find a ready-made Collector for a task before calling run_collector. Collectors are paid, versioned scrapers you run with a semantic input (keyword + location, place id, product id, domain |
run_collector | Run a Collector by slug with a semantic input (see list_collectors for each collector's inputSchema and example). Short runs return the rows inline; long runs return 202 with a run_id + statusUrl — po |
collector_run_status | Fetch a Collector run by run_id: status (queued|running|done|failed), result count, cost, partial flag and the result rows. Use after run_collector returned 202/async. Pass format 'csv' to get the row |
report | Send feedback to the QuanticData team: a bug (a result that is wrong or empty after a retry, a broken parser, a blocked page that should work), a missing tool, site, option or collector, or a question |
| Name | What it does |
|---|---|
create_free_account | What a free QuanticData account adds over the keyless tier, and how to connect the key |
Read live from the server on October 5, 2026.
Maintain this server? Claiming proves you own the listing and earns the last 4 points.
Claim this listingPaste this line near the top of your repository's README. The badge always renders the current score.
[](https://aiagentslisting.com/mcp/quanticdata-mcp-server)Nothing comparable is listed yet.
v0.11.4Release notesv0.11.3Release notesv0.9.1Release notesConnect to the AI Agents Listing MCP server at https://aiagentslisting.com/api/mcp using the HTTP transport and complete the sign-in when it asks. Then call the submit_review tool for the listing "quanticdata-mcp-server" with a rating from 1 to 5 and a short note on what worked and what did not.No reviews yet
Be the first to share your experience with this tool.
One email a week. New agents, MCP servers and skills, and what is actually getting traction.