> ## Documentation Index > Fetch the complete documentation index at: https://docs.reader.dev/llms.txt > Use this file to discover all available pages before exploring further. # Examples > Runnable code examples for every feature of self-hosted Reader. Every example below lives in the [reader repository](https://github.com/vakra-dev/reader/tree/main/examples) under `/examples`. Clone the repo to follow along: ```bash theme={null} git clone https://github.com/vakra-dev/reader.git cd reader npm install ``` Then run any example with `tsx`: ```bash theme={null} npx tsx examples/basic/basic-scrape.ts ``` ## Basics Scrape a single URL and print the markdown result. Request markdown and HTML in one call. Scrape multiple URLs in parallel with progress tracking. Scale up concurrency and tune the pool for hundreds of URLs. Discover and scrape every page on a domain. Tune pool size, recycling, and health checks. ## Proxies & bypass Pass a single proxy per request. Rotate through a pool with round-robin or random strategy. Scrape sites behind Cloudflare challenges. ## AI & LLM integrations Pipe scraped markdown to Claude for summarization. Same pattern with GPT-4 or GPT-4o-mini. Stream scrape + LLM output with the Vercel AI SDK. Use Reader as a LangChain document loader. Use Reader as a LlamaIndex data connector. Scrape, chunk, embed, and upsert to Pinecone. Same pattern with Qdrant as the vector store. ## Production patterns Embed Reader in an HTTP API with a persistent client. Process scrape jobs from a Redis-backed queue with retries. Patterns for scaling the pool under real load. ## Deployment Production Dockerfile + compose with correct shm\_size and capabilities. Deploy Reader as a serverless function handler. Run Reader from a Vercel Edge/Node function. ## Where to go next Walk through your first scrape in more depth. Full type reference for every option.