# URL Metadata Scraper MCP for AI Agents AI Agent Connect

> URL Metadata Scraper. Stop wasting tokens on full website scrapes. This Connector pulls clean OpenGraph and SEO tags (title, description, image) from any link instantly. It is built for speed and precision, giving your agent exactly what it needs for link previews and social posts without the overhead of a full browser.

## Overview
- **Category:** developer-tools
- **Price:** Free
- **Endpoint:** https://edge.vinkius.com/vk_preview_8dCiPY0j2J0NrFWar9hvAPxfBBUFBYDVdWIuQXWE/ai-agent-connect
- **Tags:** opengraph, metadata-extraction, seo-optimization, token-optimization, html-parsing, link-preview

## Description

URL Metadata Scraper lets you pull clean OpenGraph and SEO tags from any link without loading entire websites. When you ask an agent to summarize a link for a social post or a marketing dashboard, the usual approach is for the agent to grab the entire HTML of the page. That is a massive waste of context window space and usually results in the AI getting distracted by navigation menus, footers, and scripts. This Connector changes that by grabbing only the specific metadata tags, like the OpenGraph title or the primary SEO image, that actually matter for the task. It uses Cheerio to parse the data quickly in an edge environment, which means your agent gets a clean, structured snippet of information instead of a mountain of junk. It is a practical way to keep your workflows fast and your costs down. If you are managing a large catalog of links or building an automated content pipeline, having this in your Vinkius library makes it much easier to handle high-volume link processing without hitting token limits.

## Tools

### scrape_url
Pass a URL to get structured metadata like titles and images without a headless browser. This is perfect for quick link previews and SEO audits.

## Prompt Examples

**Prompt:** 
```
Check the preview info for `https://example.com`.
```

**Response:** 
```
I've pulled the metadata for that link:

*   **Title:** Example Site
*   **Description:** The best place for high-quality content and tools.
*   **Image:** `https://example.com/og.jpg`
```

**Prompt:** 
```
What's the SEO description for this link?
```

**Response:** 
```
The SEO description for that page is: *"Discover the best tools for productivity and automation."*
```

**Prompt:** 
```
Get the image for this article so I can post it.
```

**Response:** 
```
I found the main image for the article: `https://news.com/cover.jpg`.
```

## Capabilities

### Extract OpenGraph tags
Get the title, description, and image for social sharing.

### Pull SEO metadata
Retrieve the primary page title and meta description for search results.

### Save context tokens
Avoid loading full HTML pages into your agent's memory.

### Parse HTML with Cheerio
Process website data quickly without a heavy headless browser.

### Generate link previews
Create clean summaries for marketing dashboards or social threads.

## Use Cases

### Automated Social Media Posts
A social media manager wants to turn a news link into a tweet. The agent uses scrape_url to get the correct image and title for a preview.

### Bulk SEO Audits
An SEO specialist needs to check 50 landing pages. The agent pulls the meta descriptions for all of them to check for consistency.

### Link Summary Dashboards
A developer is building a dashboard that summarizes links. The agent grabs the metadata to show a clean card view of each site.

### Content Verification
A content team wants to verify that their new blog posts have the right OG tags. The agent checks the URL and reports the data.

## Benefits

- Lower costs because scrape_url only pulls the tags you need, saving your token budget on unnecessary HTML.
- Faster responses since it does not need to boot a headless browser to see the page content.
- Cleaner data for your agent because it ignores nav bars and footers to focus on the OpenGraph info.
- Accurate link previews for social media because it pulls the specific image and title tags intended for sharing.
- Easier SEO audits because you can quickly check the description and title of dozens of URLs in one go.

## How It Works

The bottom line is you get clean, usable link data without the cost or noise of a full web scrape.

1. Provide the agent with a specific URL.
2. The Connector fetches and parses the page's metadata tags.
3. You get a structured JSON object with the title, description, and image URL.

## Frequently Asked Questions

**Does the URL Metadata Scraper work for all websites?**
Yes, it works on any website that has standard meta tags. If a site doesn't have these tags, the agent will report that no data was found.

**Can I use URL Metadata Scraper to see the actual page content?**
No, this Connector is specifically designed to grab summary tags like the title and description. It does not pull the full body text of the page.

**How does URL Metadata Scraper save me money?**
It prevents your AI agent from reading the entire HTML of a page. By only pulling a few tags, you save a significant amount of tokens on every request.

**Will URL Metadata Scraper find the correct image for Twitter?**
Yes, it specifically looks for OpenGraph image tags, which are the standard tags used by Twitter and other social platforms for link previews.

**Is URL Metadata Scraper faster than a browser?**
Yes, it is much faster. It uses a lightweight parser to grab data without having to load styles, scripts, or images in a full browser window.

**Does it run a headless browser?**
No, it uses a lightweight fetch to grab the raw HTML, making it incredibly fast.

**What fields does it extract?**
Title, description, featured image URL, and site name.

**Can it bypass captchas?**
No, this is a lightweight HTTP scraper. It will fail on sites protected by Cloudflare or advanced bot protection.