webclaw is unclaimed —

webclaw

The web scraper API your AI agent deserves.·webclaw.io

Webclaw is a Rust-based web extraction toolkit that turns any URL into clean markdown, JSON, plain text, or LLM-optimized output, with automatic bot-protection bypass and JavaScript rendering handled automatically. It ships as a hosted REST API, a CLI, and an MCP server exposing 14 tools for AI agents such as Claude, Cursor, and Windsurf. It offers a Firecrawl-compatible /v2 endpoint, web search, crawling, structured LLM extraction, lead enrichment, brand extraction, and content-change tracking, and its core engine is open source and can be self-hosted.

What it's for

Stock/price monitoring across retailersTravel price monitoring (flights, hotels, packages)Lead enrichment (founders, LinkedIn, X from a company URL)Giving AI agents live web access via MCPRAG pipeline content ingestionDeep/multi-source research with citationsCompetitive intelligenceBrand extractionContent change monitoringWebsite-to-Markdown conversionYouTube transcript extraction

Features 19

9 output formats: markdown, plain text, JSON, LLM-optimized, screenshot, links, rawHtml, attributes, query
Automatic bot-protection bypass and JavaScript rendering
BFS same-origin crawler with sitemap.xml and robots.txt discovery
Brand extraction (colors, fonts, logo URL, favicon)
Browser actions (click, type, scroll, wait, press, executeJavascript) and screenshots
Content change tracking via page diffing
CSS selector filtering to include/exclude content
Document parsing (DOCX, XLSX, CSV to markdown)
Fast extraction of static pages (around 118ms)
Firecrawl v2 API-compatible endpoints
Lead enrichment (founders/leadership with LinkedIn and X from a company URL)
LLM integration chain (Ollama, then OpenAI, then Anthropic) for schema extraction, prompt extraction, and summarization
MCP server exposing 14 tools for AI agents
PDF text extraction
Proxy support via proxies.txt
Web scraping REST API (scrape, crawl, search, extract, map, batch, summarize, research, brand, diff, lead endpoints)
Web search with optional parallel scraping of results
Webhooks with Discord/Slack formatting, HMAC-SHA256 signed payloads
YouTube transcript extraction

At a glance

free tierfree trialopen sourceself-hostableAPI

TypeWeb app

DeploymentHybrid

PlatformsWeb, Windows, macOS, Linux

Fordevelopers, AI agent builders, open-source builders

Companywebclaw

Integrations

Claude DesktopClaude CodeCursorWindsurfOpenCodeCodexAntigravityLangChainn8nSerperOllamaOpenAIAnthropicDiscordSlackFirecrawl

Resources

Pricing

Signing in gives 3 free runs a day with no card required. Credits are consumed at different rates per operation: plain page 1 credit, heavy render +2, protected site +9, search/10 results 2, summarize 10, brand 5, diff 2, LLM extract 25 credits. Research runs are metered separately per month with a per-tier cap on max sources.

Starter$19/mo

Card required · cancel anytime

10,000 credits/mo, 3 research runs/mo, max 10 sources, concurrency 5

Growth$49/mo

Marked 'Popular' on the pricing page

100,000 credits/mo, 10 research runs/mo, max 20 sources, concurrency 20

Pro$99/mo

250,000 credits/month, 20 research runs/mo, Max sources 30, Concurrency 50, Priority support

250,000 credits/mo, 20 research runs/mo, max 30 sources, concurrency 50

Scale$399/mo

1,000,000 credits/month, 60 research runs/mo, Max sources 100, Concurrency 100, Priority + Slack support

1,000,000 credits/mo, 60 research runs/mo, max 100 sources, concurrency 100

DedicatedQuote

Unlimited pages, unlimited research, 200 concurrent

Open sourceFree

Self-host forever

No limits on your own hardware

Save 20% on any yearly plan