domscout - Turn any webpage into clean Markdown, screenshots & DOM data

by•
domscout is a serverless browser engine built for AI workflows and agents. Render dynamic JavaScript pages in real Chromium and get clean LLM-ready Markdown, structured DOM data, screenshots, or PDFs in a single call. Use it via REST API or instantly connect our 10 MCP tools to Claude, Cursor, or VS Code. Every request reports exact credit costs and remaining balance. Includes scheduled visual monitors, webhooks, and 100 free requests/mo (no card required).

Add a comment

Replies

Best
Maker
📌
Hey Product Hunt! 👋 I'm excited to share domscout with you today. ### Why domscout was built: If you've ever built AI agents, RAG pipelines, or automated web workflows, you know the frustration: - Simple HTTP fetches fail on modern JavaScript-heavy SPAs. - Raw HTML dumps tens of thousands of messy, noisy tokens into your LLM context window. - Hosting, scaling, and patching headless Chromium clusters in production is an infrastructure nightmare. - Most browser APIs hide their costs until you get an unpleasant billing surprise at the end of the month. We built domscout to be the browser your AI agent doesn't have. ### What makes it different: 1. Real Chromium, Clean Extraction: Every page renders in headless Chromium and is transformed into token-dense, clean Markdown and structured DOM nodes ready for prompt ingestion. 2. Visual Proof: Alongside clean text, get PNG, WebP, or PDF screenshots to verify what the page actually looked like. 3. Native MCP Support: Connect 10 browser tools directly into Claude Code, Cursor, or Claude Desktop with a single command (`npx -y /mcp`). 4. Radical Pricing Transparency: Every API response reports exactly what the call cost and what quota remains. No surprise overages. 5. Scheduled Monitors & Webhooks: Track visual and semantic changes on target pages over time with automated webhook delivery. ### Try it out! Every account gets 100 free requests/month with no credit card required. You can test live URLs instantly in our interactive playground: I’d love to hear your feedback, answer any questions, and learn what you're building with agents and browser automation; and all in all how to improve the app :)