Developer Hub
Docs, guides, SDKs, tools, and integrations in one place. Official resources are listed here today. Community resources land in their own track as developers ship them.
Get to first value
Start Here
Three ways to make your first successful call, pick the one that matches how you work.
Quickstart (curl)
One terminal, one API key, a visible response. The fastest path regardless of your stack.
QuickstartSDK Quickstart
Install, configure, first call in TypeScript or Python, shown side by side.
QuickstartMCP Server
Connect Tabstack to Claude Code, Cursor, or Windsurf and call it from inside your editor.
Setup pathUnderstand the product
Documentation
Concepts and reference. How extract, generate, research, and automate work, and when to reach for each.
Extract
How a page maps to your schema, and when structured extraction beats raw HTML parsing.
Generate
Transform or infer data the page does not already contain. How it differs from extract.
Research
How one call returns a multi-source answer with citations attached, and when to choose it over extract.
Automate
The browser agent and orchestrator model for multi-turn interactive sessions.
API Reference
Every endpoint, parameter, and response shape, generated from the spec.
Solve a real task
Guides & Recipes
Task focused how-tos and full cookbook recipes you can copy into a project.
Extract JSON
Pull structured JSON from any URL with a schema.
Markdown Extraction
Turn a page into clean markdown for prompts and pipelines.
Generate JSON
Transform or infer fields a page does not already contain into a typed schema.
Research With Citations
One call returns a multi-source answer with sources attached.
LangChain Integration
Use Tabstack as a loader and tools inside a LangChain app.
Price Monitor
Check competitor pages on a schedule and write changes to a store.
Job Listings
Extract structured job listings from multiple company career pages into one schema.
Build the integration
SDKs & Code
Typed SDKs and a worked first app you can clone and run.
Work faster
Tools
Local and editor tools that put Tabstack one command away.
MCP Server
Expose Tabstack as tools to any MCP-aware coding assistant.
Tabstack CLI
Call extract, research, and automate from your terminal. Install with a single command.
Raycast Extension
Run a quick extract or research from Raycast. Installable now from the store PR while it is in review.
Drop into your stack
Integrations
First party adapters for the frameworks agent builders already use.
LangChain TypeScript Package
The @tabstack/langchain package on npm: native LangChain.js tools for extraction, research, generation, and automation.
LangChain Python Package
The langchain-tabstack package on PyPI: a loader, tool wrappers, and a tools factory.
LangChain Adapter
Native LangChain tools for web extraction, research, generation, and automation, for both LangChain.js and Python.
Vercel AI SDK Adapter
Typed tool() definitions that pass straight to generateText and streamText, backed by the Tabstack SDK.
OpenAI Agents Adapter
Pre-built tool() definitions you hand straight to an Agent, backed by the Tabstack SDK.
Claude Agent Adapter
An in-process MCP server you wire into query(), giving Claude Agent SDK agents web access.
Mastra Adapter
Typed createTool definitions that spread straight into a Mastra Agent, backed by the Tabstack SDK.
LlamaIndex Adapter
Pre-built tool() definitions you pass straight to a LlamaIndex.TS agent, backed by the Tabstack SDK.
Pi Agent Adapter
A drop-in Pi extension that registers all five Tabstack tools with the pi CLI.
eve Adapter
Pre-built defineTool definitions you re-export into agent/tools/ to give eve agents web access.
Hermes Agent Adapter
A Hermes plugin that adds a tabstack toolset plus a Tabstack backend for Hermes' own web_extract tool.
Skip the schema writing
Extraction Schemas
Ready-made JSON Schemas for common data shapes, grouped by industry. Copy one into an extract call and get those exact fields back.
B2B Intelligence
Track what a company ships, charges, and announces. Eight schemas: company profile, competitor profile, funding round, SaaS pricing page, tech stack, product changelog entry, press release, and API documentation endpoint.
Real Estate
Pull listings and sale history off any property site into one shape. Six schemas: residential listing, commercial property, rental listing, short-term rental, property sold history, and foreclosure auction listing.
E-commerce
Monitor prices, stock, and reviews across storefronts and marketplaces. Five schemas: product listing, product review, category search results, marketplace seller profile, and grocery CPG product.
Jobs and Hiring
Read hiring signals off career pages and announcements. Five schemas: job posting, salary data point, company hiring velocity, executive leadership change, and layoff event.
Finance
Turn filings, deal pages, and asset listings into typed records. Five schemas: public company filing, crypto asset, angel and seed deal, VC portfolio page, and alternative asset listing.
Government and Public Records
Extract public records from county and agency portals that were never built for APIs. Five schemas: business registration, building permit, property tax record, court filing case, and government contract award.
Healthcare
Structure regulated data from registries and directories. Four schemas: clinical trial, provider directory entry, drug formulary entry, and medical device recall.
Travel
Capture rates and availability that change by the hour. Four schemas: hotel listing, flight price snapshot, restaurant listing, and attraction experience.
Social and Content
Normalise articles, profiles, and reviews for feeds and RAG pipelines. Three schemas: news article, social profile, and software review.
Developer Ecosystem
Read repo and package metadata for dependency and adoption tracking. Two schemas: GitHub repository and package registry entry.
For coding agents
AI Skills
Skills and shareable schemas that teach an agent to use Tabstack.
Evaluating options
Compare
Honest side-by-side write-ups against the tools developers weigh Tabstack against.
Tabstack vs Firecrawl
Where each wins for markdown and structured extraction.
Tabstack vs Browserbase
Managed extraction against managed browser infrastructure.
Tabstack vs Playwright
A managed API against running your own browser.
Tabstack vs Browser Use
Browser Use orchestrates agent workflows, Tabstack handles the structured web-intelligence calls inside them.
Tabstack vs LangChain
LangChain frames the agent, Tabstack is the extraction and research call you wire into it.
Tabstack vs Apify
A broad crawler platform with an actor marketplace against a single structured-extraction call.
Tabstack vs Stagehand
Stagehand drives a browser with AI actions, Tabstack returns structured data from a URL with no browser to run.
Tabstack vs Steel
Steel runs managed headless browsers for agents. Tabstack returns structured data from a URL with no browser to run.
Tabstack vs Exa
Exa is a neural search API for finding pages. Tabstack extracts structured data and cited research from them.
Tabstack vs fastCRW
How Tabstack compares to fastCRW for turning pages into structured data.
Tabstack vs Parallel AI
Parallel AI is a web research API for agents. Tabstack covers research plus structured extraction in one API.
Tabstack vs Bright Data
Bright Data is large-scale proxy and scraping infrastructure. Tabstack is a single structured-extraction API call.
Tabstack vs Perplexity
Perplexity is a consumer AI answer engine. Tabstack is the research and extraction API you build into your own product.
Built by developers
Community
Resources built by developers using Tabstack, hand-picked by the team. A separate track from the official docs above.
These are not official Tabstack resources, so support and maintenance are up to their authors.
Start building with Tabstack.
Get a structured response from any URL in one call. No browser to run, no parser to maintain. Free to start.