AgentQL

aid: agentql
name: AgentQL
description: AgentQL connects LLMs and AI agents to the entire web through a specialized query language, REST API, and Python/JavaScript
  SDKs. It enables web scraping, data extraction, and browser automation using natural language queries that are self-healing
  — adapting automatically to page layout changes. AgentQL supports structured data extraction from web pages, PDF documents,
  and images, and integrates with LangChain, LlamaIndex, MCP, Zapier, and Google ADK.
url: https://raw.githubusercontent.com/api-evangelist/agentql/refs/heads/main/apis.yml
humanURL: https://www.agentql.com/
type: Index
image: https://kinlane-productions.s3.amazonaws.com/apis-json/apis-json-logo.jpg
access: 3rd-Party
position: Consumer
tags:
- Agents
- Artificial Intelligence
- Web Scraping
- Data Extraction
- Browser Automation
- REST API
created: '2025-08-19'
modified: '2026-05-19'
specificationVersion: '0.19'
apis:
- aid: agentql:query-data-api
  name: AgentQL Query Data API
  description: Extract structured JSON data from web pages using AgentQL query language or natural language prompts. Supports
    URL-based and raw HTML input, configurable browser behavior, proxy settings, and screenshot capture.
  humanURL: https://docs.agentql.com/rest-api/api-reference
  baseURL: https://api.agentql.com
  tags:
  - Data Extraction
  - Web Scraping
  - REST API
  - AI
  properties:
  - type: Documentation
    url: https://docs.agentql.com/rest-api/api-reference
  - type: GettingStarted
    url: https://docs.agentql.com/quick-start
  - type: Authentication
    url: https://docs.agentql.com/rest-api/authentication
  - type: OpenAPI
    url: https://raw.githubusercontent.com/api-evangelist/agentql/refs/heads/main/openapi/agentql-openapi.yaml
  - type: NaftikoCapability
    url: capabilities/agentql-query-data.yaml
  - type: NaftikoCapability
    url: capabilities/agentql-query-document.yaml
  - type: NaftikoCapability
    url: capabilities/agentql-remote-browser.yaml
- aid: agentql:remote-browser-api
  name: AgentQL Remote Browser Sessions API
  description: Create and manage remote Chrome browser sessions with Chrome DevTools Protocol (CDP) access for authenticated
    web automation, stealth browsing, and complex multi-step interactions.
  humanURL: https://docs.agentql.com/rest-api/api-reference
  baseURL: https://api.agentql.com
  tags:
  - Browser Automation
  - Remote Browser
  - CDP
  - Web Automation
  properties:
  - type: Documentation
    url: https://docs.agentql.com/rest-api/api-reference
- aid: agentql:query-document-api
  name: AgentQL Query Document API
  description: Extract structured data from PDF documents and images (JPEG, PNG) using AgentQL query language or natural language
    prompts. Useful for processing invoices, reports, and other document formats.
  humanURL: https://docs.agentql.com/rest-api/api-reference
  baseURL: https://api.agentql.com
  tags:
  - Document Parsing
  - PDF
  - Data Extraction
  - AI
  properties:
  - type: Documentation
    url: https://docs.agentql.com/rest-api/api-reference
common:
- type: Portal
  url: https://www.agentql.com/
- type: Documentation
  url: https://docs.agentql.com/home
- type: GettingStarted
  url: https://docs.agentql.com/quick-start
- type: Pricing
  url: https://www.agentql.com/pricing
- type: Blog
  url: https://www.agentql.com/blog
- type: Support
  url: https://docs.agentql.com/support
- type: Console
  url: https://dev.agentql.com/playground
- type: SDK
  url: https://pypi.org/project/agentql/
  title: Python SDK
- type: SDK
  url: https://www.npmjs.com/package/agentql
  title: JavaScript SDK
- type: CLI
  url: https://docs.agentql.com/cli
- type: GitHubOrganization
  url: https://github.com/tinyfish-io
- type: Features
  data:
  - name: Natural Language Query Language
    description: A specialized query language that uses natural language to locate and extract web elements without requiring
      XPath, CSS selectors, or regex.
  - name: Self-Healing Queries
    description: AI-powered queries automatically adapt to page layout changes, eliminating brittle scrapers that break on
      site updates.
  - name: REST API
    description: Browserless data extraction from public URLs via a REST API requiring only an API key and query parameters.
  - name: PDF and Image Parsing
    description: Extract structured data from PDF documents, JPEG, and PNG images using the same query language as web extraction.
  - name: Remote Browser Sessions
    description: Managed Chrome browser sessions with CDP access for authenticated browsing, stealth mode, and complex multi-step
      web automation.
  - name: Playwright Integration
    description: Python and JavaScript SDKs extend Playwright with AgentQL query capabilities for AI-powered browser automation.
  - name: Browser Debugger Extension
    description: Chrome extension for real-time query testing and optimization during development.
- type: UseCases
  data:
  - name: E-Commerce Price Monitoring
    description: Extract product names, prices, and availability from e-commerce sites for competitive intelligence and price
      tracking.
  - name: Job Board Aggregation
    description: Collect job listings, requirements, and company information from multiple job boards into a unified dataset.
  - name: Social Media Content Harvesting
    description: Extract posts, metrics, and profile data from social media platforms for analysis and reporting.
  - name: Document Data Extraction
    description: Parse invoices, contracts, and reports in PDF format to extract structured data for downstream processing.
  - name: AI Agent Web Access
    description: Enable AI agents to access and extract data from any website as part of automated research and task completion
      workflows.
  - name: Lead Generation
    description: Automate the collection of contact information, company data, and other business intelligence from public
      web sources.
- type: Integrations
  data:
  - name: LangChain
    description: Integrate AgentQL web extraction into LangChain agent and chain workflows.
  - name: LlamaIndex
    description: Use AgentQL as a data ingestion source for LlamaIndex-powered RAG and agent applications.
  - name: MCP (Model Context Protocol)
    description: Expose AgentQL capabilities as MCP tools accessible to any MCP-compatible AI agent.
  - name: Zapier
    description: Connect AgentQL web extraction to thousands of apps via Zapier automation workflows.
  - name: Google ADK
    description: Integrate AgentQL with Google Agent Development Kit for Gemini-based agent web access.
  - name: Langflow
    description: Use AgentQL in Langflow visual AI workflow pipelines for web data extraction.
  - name: Playwright
    description: Extend Playwright browser automation with AgentQL AI-powered element querying and data extraction.
- type: Integrations
  url: https://www.agentql.com/integrations
- name: MCP Server
  url: https://github.com/tinyfish-io/agentql-mcp
  type: MCPServer
integrations:
- name: AgentQL
maintainers:
- FN: Kin Lane
  email: [email protected]