> ## Documentation Index
> Fetch the complete documentation index at: https://web2md.org/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Custom Extraction Rules

> Define per-domain CSS selectors for precise content extraction.

## Overview

Custom Extraction Rules (v0.7) let you define per-domain CSS selectors that control exactly what Web2MD extracts. If a site's main content lives in an unusual container, or if there are elements you always want stripped, custom rules give you full control.

<Note>
  Custom Rules is a **Pro-only** feature.
</Note>

## Creating a rule

<Steps>
  <Step title="Open Settings">
    In the Web2MD popup, navigate to **Settings → Custom Rules**.
  </Step>

  <Step title="Add a new rule">
    Click **New Rule** and enter the domain (e.g., `example.com`).
  </Step>

  <Step title="Define your selectors">
    Set the content selector, remove selectors, and optionally a title selector.
  </Step>

  <Step title="Save and convert">
    Save the rule. Next time you convert a page on that domain, your custom selectors will be applied automatically.
  </Step>
</Steps>

## Rule fields

| Field                | Description                                           | Example                          |
| -------------------- | ----------------------------------------------------- | -------------------------------- |
| **Domain**           | The domain this rule applies to (exact match)         | `docs.example.com`               |
| **Content selector** | CSS selector for the main content container           | `article.post-body`              |
| **Remove selectors** | CSS selectors for elements to strip (comma-separated) | `.ads, .sidebar, .related-posts` |
| **Title selector**   | Override automatic title detection                    | `h1.entry-title`                 |

## Example: cleaning up a blog

Suppose you regularly convert articles from a blog that wraps content in a specific div and includes noisy sidebar widgets:

```
Domain:           blog.example.com
Content selector: div.article-content
Remove selectors: .author-bio, .newsletter-signup, .comments-section
Title selector:   h1.post-title
```

With this rule, Web2MD will:

1. Extract only the content inside `div.article-content`
2. Strip the author bio, newsletter signup, and comments
3. Use the `h1.post-title` element as the document title

## How rules are applied

<CardGroup cols={2}>
  <Card title="Automatic matching" icon="bullseye">
    Rules are matched by domain. When you convert a page, Web2MD checks if a custom rule exists for that domain and applies it before the standard extraction pipeline.
  </Card>

  <Card title="Fallback behavior" icon="shield">
    If your content selector doesn't match anything on the page, Web2MD falls back to its default extraction logic. Your rule won't break conversions.
  </Card>
</CardGroup>

## Managing rules

All rules are managed in **Settings → Custom Rules** within the extension popup.

| Action     | How                                                          |
| ---------- | ------------------------------------------------------------ |
| **Create** | Settings → Custom Rules → New Rule                           |
| **Edit**   | Click any existing rule to modify its selectors              |
| **Delete** | Click the trash icon next to a rule                          |
| **Test**   | Navigate to a page on the target domain and run a conversion |

<Tip>
  Use your browser's DevTools (`Cmd+Shift+I` or `F12`) to inspect the page and find the right CSS selectors before creating a rule.
</Tip>

## Community sharing

<Info>
  Community rule sharing is coming soon. You'll be able to mark rules as public (`isPublic` flag) so other users can benefit from your domain-specific selectors.
</Info>

## Relationship to Site Adapters

Custom Rules and [Site Adapters](/docs/advanced/site-adapters) serve different purposes:

|                   | Site Adapters                                       | Custom Rules               |
| ----------------- | --------------------------------------------------- | -------------------------- |
| **Scope**         | Built-in, specific platforms (Reddit, GitHub, etc.) | Any domain you define      |
| **Method**        | Uses platform APIs and specialized logic            | CSS selectors only         |
| **Maintained by** | Web2MD team                                         | You                        |
| **Priority**      | Runs first                                          | Runs if no adapter matches |

If a Site Adapter exists for a domain, it takes priority over custom rules. Custom rules are best for sites without a dedicated adapter.