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

> ## Agent Instructions
> Setting up Context.dev for a user? Follow https://docs.context.dev/agent-quickstart.md. MCP clients sign in with OAuth and need no API key; application code reads CONTEXT_DEV_API_KEY.

# Content filtering

> Select the page regions used by text and image outputs.

Use `sharedParams` to keep useful content and remove navigation, footers, or repeated blocks. Filters affect HTML, Markdown, images, parsed fields, highlights, and JSON.

## Keep the article

Send this body to `POST /web/scrape`. See the [Quickstart](/quickstart) for authentication and SDK setup.

```json theme={null}
{
  "url": "https://example.com/article",
  "formats": {
    "markdown": true,
    "images": true
  },
  "sharedParams": {
    "includeSelectors": [
      "article"
    ],
    "excludeSelectors": [
      ".newsletter",
      ".advertisement"
    ]
  }
}
```

## Options

| Field              | Default | Effect                                                |
| ------------------ | ------- | ----------------------------------------------------- |
| `mainContentOnly`  | `false` | Strip detected page chrome and keep the main content. |
| `includeSelectors` | None    | Keep matching subtrees.                               |
| `excludeSelectors` | None    | Remove matching subtrees; exclusions win.             |
| `includeFrames`    | `false` | Include iframe content in HTML and text extraction.   |

Selector lists contain 1–50 unique selectors, each up to 2,048 characters. Use CSS; XPath is unsupported. For a gallery, set `includeSelectors` to its container and request `images`.

## Which outputs change

Screenshots, original bytes, product extraction, and metadata headings use the page independently of these filters. Screenshots show visible frames even when `includeFrames` is false.

If a selector matches nothing, the output can succeed with an empty string or list. Inspect the page’s HTML and adjust the selector before retrying. [Parse fields](/scrape/parse-fields) rules run on the filtered HTML.