Docs to Markdown and llms.txt
Pricing
from $2.00 / 1,000 page converteds
Docs to Markdown and llms.txt
Turn a documentation site into clean Markdown pages plus ready llms.txt and llms-full.txt files for RAG, AI agents and offline search.
Pricing
from $2.00 / 1,000 page converteds
Rating
0.0
(0)
Developer
Meta Parse
Maintained by CommunityActor stats
0
Bookmarked
2
Total users
1
Monthly active users
3 days ago
Last modified
Categories
Security
Limited permissionsShare
Turn a documentation site into clean Markdown pages, plus ready-made llms.txt and llms-full.txt files. Use it to feed RAG pipelines and coding agents, or to keep an offline copy of the docs you rely on.
Navigation, sidebars, footers and page chrome are removed. Code blocks stay fenced, and links are made absolute.
Input
{"startUrl": "https://www.python-httpx.org/","maxPages": 50,"siteTitle": "HTTPX"}
Only pages under the start path are converted. Starting at https://example.com/docs/ won't pull in the marketing site or blog. Pages are found through the sitemap and by following links.
Output
One record per page:
{"url": "https://www.python-httpx.org/async/","title": "Async Support","description": null,"markdown": "# Async Support\n\n…","wordCount": 785,"likelyClientRendered": false,"headings": ["Async Support", "Making Async requests"]}
The run's key-value store also holds:
llms.txt: an index of pages with linksllms-full.txt: every page in one file, with source comments
Pricing
$0.002 per page converted. The default limit is 50 pages, or $0.10. Raise maxPages for bigger sites: 300 pages costs at most $0.60.
Each run also has a $0.00005 start fee per GB of memory. All compute is included.
Limits
- JavaScript-only sites: there's no browser, so a site that renders only in the browser returns near-empty pages. These are flagged with
likelyClientRendered: true. Most static generators, such as MkDocs, Docusaurus and Sphinx, send full HTML. - Robots rules: it follows the site's
robots.txt. - Rights: check the licence of any documentation you redistribute.