Anakin

Anakin is a web data API from Anakin Inc (YC S21) for scraping, crawling, searching and acting on websites, with an open-source self-hostable scraping engine published under AGPL-3.0. It returns clean Markdown, HTML or structured JSON for use in AI agents, RAG pipelines and data extraction workflows.

What It Is

Anakin is a scraping and web-data platform. The hosted anakin.io service exposes products including URL Scraper, Crawl, Map, Search API, Agentic Search, Browser API, Browser Sessions, Website Monitoring, AI Visibility and Wire, which the site describes as managed APIs for sites that do not offer one. The open-source repository, AnakinScraper OSS, is a Go server that scrapes a URL and returns HTML, cleaned HTML, Markdown and optionally AI-extracted JSON.

How the Self-Hosted Engine Works

Each scrape job runs through a handler chain: a plain HTTP fetch first, then a Camoufox anti-detect Firefox browser for JavaScript-heavy pages, then an optional external API handler such as the anakin.io hosted service. Per-domain configs set handlers, timeouts, retries, headers and failure-detection patterns, and a Thompson Sampling proxy selector picks proxies per domain. Structured JSON extraction uses Gemini with the user's own API key. The server offers sync, async and batch endpoints (batches of up to 10 URLs) and a React web dashboard, and can run with just Go and in-memory storage or with Docker and PostgreSQL.

Hosted Platform and Agent Access

The vendor describes a keyless Zero Touch tier in which an agent can scrape a URL or run read-only Wire actions in a single unauthenticated HTTP call. It lists 940+ platforms in the Wire catalog, SDKs for many languages, a CLI, an MCP server, and integrations with Claude Code, Cursor, OpenClaw, Zapier, Make, n8n, Dify, LangChain, LlamaIndex, CrewAI, Langflow and Flowise. Anakin also runs managed enterprise competitive-intelligence pipelines such as price monitoring and product matching, and states it holds SOC 2 Type II and ISO 27001 certifications.