A self-hosted alternative to Firecrawl for web scraping. It converts web pages into structured Markdown to optimize token usage in LLM prompts and RAG pipelines.
Tech Stack
TypeScriptAI, ML & Data
Why reader?
• High-quality content extraction
• Reduces LLM token consumption
• Easy to self-host via Docker
Limitations
• May struggle with heavy SPA sites
• Requires proxy for geo-blocking
• Resource intensive for large crawls
3/6/2026
Last Update
769
Forks
124
Issues
Apache-2.0
License
Financial Leak Detected
Stop the "SaaS Tax"
Your team could be burning cash. Switching to reader instantly boosts your runway.
