Web-to-Markdown Converter
Convert pasted HTML or a fetched URL to Markdown for LLM input, dropping scripts, navigation and footers, and estimate the token savings.
Web-to-Markdown Converter
Output
Working…Updated . Provided as is. Check the output before you rely on it in production.
How to use Web-to-Markdown Converter
- 1
Paste HTML or a URL
Paste a page's HTML, or switch to Fetch URL mode to load a public page.
- 2
Convert
Headings, paragraphs, links and lists are converted to Markdown, with scripts and styles dropped.
- 3
Check the token savings
Compare the estimated token count of the HTML and the Markdown to see how much context you save.
- 4
Copy the Markdown
Copy it into a prompt, a RAG index or documentation.
Questions and answers
Why convert web pages to Markdown for LLMs?
Why does Fetch URL fail on some sites?
Use it as an API
Web-to-Markdown Converter is also callable as a free HTTP JSON API at
https://aidevhub.io/api/web-to-markdown/ — GET with query
parameters or POST with a JSON body, no authentication, CORS enabled, fair use
(abusive traffic is throttled at the edge; there is no per-request quota header). Responses return
{ ok, tool, result, meta }.
curl -s -X POST https://aidevhub.io/api/web-to-markdown/ \
-H "Content-Type: application/json" \
-d '{"input":"<html><body><h1>Example</h1><p>Hello world.</p></body></html>","mode":"html"}' Machine-readable contract: /api/tool/web-to-markdown.json All API endpoints: /agents/ LLM site index: /llms.txt
For AI agents: how to call this tool
Machine-readable contract, endpoints and examples. Humans can ignore this section.
Best Path For Builders
Dedicated API endpoint
Deterministic outputs, machine-safe contracts, and production-ready examples.
Dedicated API
https://aidevhub.io/api/web-to-markdown/ OpenAPI: https://aidevhub.io/api/openapi.yaml
Unified Runtime API
https://aidevhub.io/api/tools/run/?toolId=web-to-markdown&a=%3Chtml%3E%0A%20%20%3Cbody%3E%0A%20%20%20%20%3Ch1%3EExample%3C%2Fh1%3E%0A%20%20%20%20%3Cp%3EHello%20world.%3C%2Fp%3E
GET and POST are supported at /api/tools/run/ with identical validation and limits.
Limit: 10 req / 60s, input max 512 KB.