Web-to-markdown API with website discovery built in
Find every page.
Get clean markdown back.
NeoReader discovers the URLs on a website and converts each page into LLM-ready markdown, built end to end for high-throughput ingestion. REST API and MCP server.
At a glance
Web-to-markdown API with website discovery built in
- Status
- Live. Free plan available.
- Built for
- AI engineers and data teams building RAG pipelines or LLM ingestion flows
Why it matters
Positioning pillars
Website discovery
Start from a domain, not a URL list.
Point NeoReader at a site and it finds the pages: sitemap parsing, crawl depth and page limits, include and exclude patterns, path prefixes, robots.txt respected by default. Each URL comes back with its title and description, ready to convert.
Speed at ingestion scale
Under 500ms for static pages.
Built for pipelines that ingest thousands of URLs per batch. Static pages take a fast HTTP path; a browser renders only when the page demands it. Fixed per-request pricing, with JavaScript rendering included.
Framework-aware rendering
Handles React, Next, Vue.
Most sites return useful HTML on the first request, and NeoReader converts them in milliseconds. JS-heavy sites (React, Next.js, Vue, Nuxt, Angular, Svelte) fall back to browser rendering, so the markdown reflects what a user actually sees, not an empty shell.
Transparent, predictable output
Structured output, every time.
Clean markdown with title, body, and metadata in one response. Boilerplate and ads removed. No surprise HTML entities, no stale cached fragments, no tracking pixels. The output a RAG pipeline can index without a post-processing step.
The mechanics
How it works
Step 1
Discover the site
Start a discovery job on a domain. NeoReader reads the sitemap, crawls within your limits and patterns, and returns the URL list with titles and descriptions.
Step 2
Convert each URL
Send a URL to the API or call the MCP server. Receive clean markdown, title and metadata in a single response, with no manual parsing.
Step 3
Fallback only when needed
Static pages return in milliseconds. JS-heavy pages automatically fall back to browser rendering: slower, but complete. Rendering is included in the per-request price.
Step 4
Drop into your RAG pipeline
Pipe the markdown straight into a vector database, a chunker or your agent's context. NeoReader ends where your knowledge pipeline begins.
Proof in production
<500ms
Static page to clean markdown
In production since 2025 as the web fetcher inside the Neomanex knowledge library that Gnosari uses to learn from websites.
NeoReader production metrics
The ecosystem
Fits into the portfolio
Related products
Questions, answered