Feeding raw HTML to a model wastes most of the context window on navigation junk. This strips a page to the actual content as markdown — the boring step that makes everything downstream work.
Still needs you to respect robots.txt and rate limits. It's a tool, not permission.
| What you get free | Free and open source (Apache-2.0), no API key, runs locally |
| Where the limit sits | No limits self-hosted |
| Credit card to start | No |
| Pricing type | Open source |
| Last checked | 2026-07-30 · source |
| Run it yourself | Yes — Apache-2.0 · repo |
| Platforms | linux, mac, windows |
| Category | Data & backend · Code & dev · Self-hosted / local |
We recheck every entry. Subscribers hear first when a free tier quietly shrinks.