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

# Read Web Pages

> Fetch a URL as context or import web material into a knowledge base.

Use Web Fetch when you already have a URL and want the Agent to read its content. Use [Web Search](/ai/web-search) when you need to discover relevant pages.

## Read a page

1. Open **Settings → Global Providers Settings → Web Fetch** and choose a provider.
2. Paste the URL into your Agent conversation and ask a specific question about it.
3. Check the retrieved content and source before relying on the answer.

| Provider              | Use                                                                    |
| --------------------- | ---------------------------------------------------------------------- |
| Local Fetch           | Retrieve content through local fetching, with browser-assisted options |
| Jina AI Link Reader   | Retrieve readable content through Jina                                 |
| Firecrawl Link Reader | Retrieve content through Firecrawl                                     |
| Enconvo Cloud Plan    | Use the Cloud link-reading service                                     |

A provider that retrieves a public URL does not automatically have access to your signed-in browser session. For authenticated or interactive pages, use the supported [Browser Extension](/features/browser-extension) or [Browser Control](/features/browser-control) flow.

## Keep web content

Add supported URLs or crawled content to a [Knowledge Base](/ai/knowledge-base) when you want to retrieve it later. Review the imported pages and indexing status; web pages can change after import.

## Troubleshooting

For empty content, check that the page is readable without extra interaction and whether browser-assisted fetching is needed. Login screens, blocked requests, dynamically loaded content, and unsupported sites can prevent extraction. Ask the Agent to identify what it actually retrieved instead of inferring missing page contents.