massanaroger avatar

Extracto Mcp

massanaroger/extracto-mcp

Connects Claude to the Extracto API for structured web scraping. You pass a URL and a JSON schema describing the fields you want (title, price, tags, nested objects, arrays), and it returns validated, typed data with null for anything missing instead of hallucinations. Exposes four tools: synchronous extract for quick jobs under 90 seconds, async extract for heavy pages, plus get_job and list_jobs for polling. Requires an Extracto API key. Useful when you need reliable data extraction from web pages without writing parsers or fighting with LLM prompt engineering to get consistent JSON shapes.

tools
4
auth
Secret needed
transport
STDIO
env vars
3
last updated
Jun 10, 2026

Tools

4 tools

Public tool metadata for what this MCP can expose to an agent.

extractSynchronous extraction from a single URL (up to ~90s). Returns { data, meta }.

Synchronous extraction from a single URL (up to ~90s). Returns { data, meta }.

No parameter schema in public metadata yet.

extract_asyncSubmit an async job for heavy or anti-bot pages. Returns a job id immediately.

Submit an async job for heavy or anti-bot pages. Returns a job id immediately.

No parameter schema in public metadata yet.

get_jobPoll an async job for status and result.

Poll an async job for status and result.

No parameter schema in public metadata yet.

list_jobsList your recent async jobs.

List your recent async jobs.

No parameter schema in public metadata yet.

Sponsors

Configuration

EXTRACTO_API_KEY*secret

Your Extracto API key from https://app.getextracto.dev/keys

EXTRACTO_BASE_URL

Override the API host (defaults to https://app.getextracto.dev)

EXTRACTO_TIMEOUT_MS

Per-request timeout in milliseconds (default 90000)

Readme

extracto-mcp

Model Context Protocol server for Extracto. It gives Claude, Cursor, Claude Code, and any MCP client the ability to turn a URL plus a schema into validated, typed JSON — no prompt engineering, no HTML parsing, and no hallucinated fields (missing data comes back as null).

Quick start

You need an Extracto API key. Get one at app.getextracto.dev/keys.

The server runs over stdio and is published to npm, so most clients just need this config block.

Claude Desktop

Edit claude_desktop_config.json (Settings → Developer → Edit Config):

{
  "mcpServers": {
    "extracto": {
      "command": "npx",
      "args": ["-y", "extracto-mcp"],
      "env": { "EXTRACTO_API_KEY": "exa_live_your_key_here" }
    }
  }
}

Cursor

Add to ~/.cursor/mcp.json (or the project .cursor/mcp.json) with the same block.

Claude Code

claude mcp add extracto -e EXTRACTO_API_KEY=exa_live_your_key_here -- npx -y extracto-mcp

View the full README on GitHub

Details

Repository
github.com/massanaroger/extracto-mcp
Category
Data & Analytics
Package
extracto-mcp
Transport
STDIO
Auth
Required
Registry
MCP registry, active

Related Data & Analytics MCP servers

View all
1periodix avatarPeriodix Actionsnet.periodix/actionsSearch LinkedIn & Sales Navigator. Get structured people, companies, posts, jobs data. 4.9/5 on G2.0stars
2deadpixel avatarRoundtabledeadpixel/roundtable-dashboardMulti-model AI debates: GPT-4o, Claude, Gemini & 200+ models discuss, then synthesize insight.0stars
3robotfleet-hq avatarSecurity Orchestrarobotfleet-hq/security-orchestraMulti-agent MCP platform for data centers and critical power.0stars
4sgunk13 avatarKuriflow Mcpsgunk13/kuriflow-mcpMake AI tasks repeatable. Run with new data automatically — zero AI tokens per run.0stars
5sharozdawa avatarSchema Gensharozdawa/schema-genGenerate Schema.org JSON-LD markup for 12 types including Person, Product, and FAQ.0stars
6shawn5cents avatarAgentskinshawn5cents/agentskinThe Semantic Layer for AI Agents. Prunes noisy API data by 70%+ for token-efficient reasoning.0stars