Convert any web page URL into clean, LLM-ready markdown using Jina Reader (r.jina.ai)

Documented steps

1. GET https://r.jina.ai/<full-target-url> (e.g. curl "https://r.jina.ai/https://www.example.com"); works with no key at a low rate limit. Docs: https://jina.ai/reader/ and https://docs.jina.ai/.
2. Add 'Authorization: Bearer $JINA_API_KEY' (free key from https://jina.ai/) for much higher rate limits.
3. Control output with request headers (https://docs.jina.ai/): X-Return-Format (markdown|html|text|screenshot|pageshot), X-Engine (browser for best quality, direct for speed, cf-browser-rendering), X-Target-Selector / X-Wait-For-Selector / X-Remove-Selector (CSS selectors to focus, wait for, or strip elements), X-Timeout, X-Token-Budget (max tokens), X-Retain-Images: none (strip images), X-With-Generated-Alt (generated alt text for images), X-No-Cache: true (bypass cache), X-Locale, X-Set-Cookie, X-Proxy-Url.
4. Choose the response shape with the Accept header: default is plain text/markdown; 'Accept: application/json' returns JSON with url/title/content; 'Accept: text/event-stream' streams large pages incrementally.