# Diffbot MCP for AI Agents AI Agent Connect

> Diffbot MCP for AI Agents. This Connector lets your agent extract structured data from any web page, query a massive knowledge graph for company and person profiles, and automatically enrich your lead lists. Stop dealing with messy copy-pasting or broken scraping scripts.

## Overview
- **Category:** artificial-intelligence
- **Price:** Free
- **Endpoint:** https://edge.vinkius.com/vk_preview_5HEsQ3kHnUpmnDG46R4fceuP7Qbytx3PkQCCCPhX/ai-agent-connect
- **Tags:** knowledge-graph, data-extraction, machine-learning, structured-data, web-crawling, metadata-parsing

## Description

Imagine you are trying to build a lead list or research a new market. Usually, that involves opening dozens of tabs, copy-pasting company names into a spreadsheet, and hunting for firmographics across multiple websites. With this Connector, you just tell your AI agent to do the heavy lifting. It handles the messy web scraping for you, turning raw URLs into clean, structured data like product prices, article contents, or forum threads. It also taps into a huge knowledge graph so you can ask about specific industries or find company data without leaving your chat window. It works like a research assistant that knows how to parse a messy webpage and find the actual information you need. You can find it in the Vinkius catalog to get your agent connected in minutes. Instead of writing complex scraping scripts that break every time a website updates, you just describe what you need in plain English. Your agent handles the detection, the extraction, and the enrichment, giving you a high-fidelity database of the information you actually care about.

## Tools

### analyze_page_type
Automatically detects what kind of page a URL is to pick the right model. This ensures your agent always gets the best data.

### enhance_person_profile
Adds details like social profiles and job history to person records. Use this to build better professional profiles.

### extract_article_data
Pulls clean text, authors, and sentiment from news or blog posts. It saves you from digging through ads and headers.

### extract_forum_thread
Grabs comments and discussions from forums or message boards. It helps you gather community sentiment quickly.

### extract_images
Identifies and pulls the primary images from a webpage. This is great for quickly gathering visual assets.

### extract_product_data
Pulls e-commerce info like prices and descriptions from product pages. It makes competitive price tracking much faster.

### search_knowledge_graph
Queries a database of billions of entities using DQL for market signals. It finds organization and people data at scale.

### verify_api_credentials
Checks if your Diffbot API keys are working correctly. Use this to troubleshoot connection issues quickly.

### enhance_company_profile
Enriches company data with social links, employee counts, and funding info. This helps keep your lead database accurate and detailed.

### extract_video_metadata
Finds and identifies embedded videos on a page. Use this to catalog video content across different sites.

### list_active_crawls
Shows the status of your current crawling jobs. It helps you keep track of large scale data extraction tasks.

## Prompt Examples

**Prompt:** 
```
Extract the main content from the article at 'https://example.com/news'.
```

**Response:** 
```
I've processed the article for you. Here is the clean content:

**Title:** New Breakthroughs in AI
**Author:** Jane Doe
**Summary:** The article discusses recent leaps in transformer models and their impact on efficiency.

**Sentiment:** Positive

Would you like me to extract any specific data points from this, like dates or names?
```

**Prompt:** 
```
Search for companies in 'New York' with more than 500 employees.
```

**Response:** 
```
I've queried the knowledge graph and found several organizations in New York with over 500 employees. Here are some highlights:

| Company | Industry | Employee Count |
| --- | --- | --- |
| TechNova | Software | 1,200 |
| UrbanLogistics | Transport | 850 |
| GreenGrid | Energy | 600 |

Would you like the full firmographics for any of these?
```

**Prompt:** 
```
Enhance the company profile for 'TechNova' using domain 'technova.com'.
```

**Response:** 
```
I've enriched the profile for TechNova. Here is the updated information:

*   **Industry:** Software Development
*   **Headquarters:** New York, NY
*   **Employee Count:** 1,200
*   **Social Links:** [LinkedIn](https://example.com), [Twitter](https://example.com)
*   **Latest Funding:** $50M Series C

Shall I look for the key people at this organization next?
```

## Capabilities

### Extract structured data from any URL
Turn messy webpages into clean data like product details or article text instantly.

### Query a global knowledge graph
Search a database of billions of entities to find company firmographics and market signals.

### Enrich lead profiles automatically
Add social links, employee counts, and funding data to your existing company or person records.

### Auto-detect page types
The agent identifies the type of URL to apply the most accurate extraction model automatically.

### Monitor crawling jobs
Check the status of your active crawls and monitor your account usage in real time.

## Use Cases

### Automated Lead Enrichment
A growth marketer asks the agent to find the LinkedIn profiles and employee counts for 50 competitors. The agent uses enhance_company_profile to fill the sheet.

### Deep Market Research
A researcher asks for a list of AI companies in San Francisco with over 1000 employees. The agent uses search_knowledge_graph to pull the data.

### Content Aggregation
A content manager asks the agent to pull the main text and sentiment from 20 different news articles. The agent uses extract_article_data for each.

### E-commerce Price Tracking
A product manager asks the agent to grab prices and descriptions from a list of competitor product pages. The agent uses extract_product_data.

## Benefits

- Stop manual scraping. Use extract_article_data and extract_product_data to get clean data from any URL instantly.
- Get deeper insights. Use search_knowledge_graph to find firmographics and market signals without searching multiple sites.
- Build better lead lists. Use enhance_company_profile and enhance_person_profile to automatically fill in missing data.
- Save on dev time. Don't write brittle scraping scripts; let the Connector handle page detection with analyze_page_type.
- Keep track of everything. Use list_active_crawls to see your progress and verify_api_credentials to keep things running.

## How It Works

The bottom line is that it turns your AI into a hands-on data engineer that handles the scraping and research for you.

1. Get your API token from your Diffbot dashboard.
2. Connect this Connector to your AI client through the Vinkius catalog.
3. Ask your agent to pull data from a URL or query the knowledge graph for specific entities.

## Frequently Asked Questions

**How does the Diffbot MCP help with market research?**
It gives your agent access to a massive knowledge graph. You can ask for specific industry signals, company firmographics, and people profiles without having to search multiple websites manually.

**Can I use the Diffbot MCP to pull data from e-commerce sites?**
Yes. The Connector can extract product data like prices, descriptions, and specifications from various online stores into a structured format for you.

**Will the Diffbot MCP work on any website?**
It is designed to handle a wide variety of URLs. It automatically detects the page type to ensure it uses the best method for extracting articles, products, or forum threads.

**How does the Diffbot MCP handle lead enrichment?**
It can take a company name or domain and automatically find missing details like employee counts, social links, and funding history to keep your records up to date.

**Is the Diffbot MCP good for scraping news content?**
It is excellent for this. It pulls clean text from news and blog posts while stripping away the noise like ads and navigation menus.

**Can I monitor my data extraction jobs with the Diffbot MCP?**
Yes. You can ask your agent to check the status of active crawls and monitor your API usage to keep your workflows running smoothly.

**How do I find my Diffbot API Token?**
Log in to your Diffbot account and navigate to the **Dashboard** or **Manage Tokens** section to copy your unique access token.

**What is DQL and how can I use it?**
DQL (Diffbot Query Language) allows you to filter the Knowledge Graph. Use the `search_knowledge_graph` tool with queries like `type:Organization industries:"AI"`.

**Can I extract comments from articles?**
Yes! The `extract_article_data` tool has an optional `discussion` parameter. Set it to `true` to retrieve structured comment threads if available.