# research-source-count is a research analysis MCP. AI Agent Connect

> research-source-count lets your AI client organize and quantify research materials. It breaks down large datasets into specific categories like Academic, Media, or Technical, so you can see exactly what your source distribution looks like without manual counting.

## Overview
- **Category:** productivity
- **Price:** Free
- **Endpoint:** https://edge.vinkius.com/vk_preview_5h0UMnEQRgDA0Pe6uurTbNc2VAnVSTLybm8e6mMh/ai-agent-connect
- **Tags:** research, data, counting, categorization, analysis

## Description

If you are working with massive piles of research data, you know how tedious it is to manually tally up how many papers, articles, or technical docs you actually have. This MCP handles that heavy lifting for you. You can hand a list of sources to your AI client and immediately get a breakdown of what you're looking at. It validates that your data follows the right schema, filters out the noise to find specific types, and provides a clear view of your source distribution. Instead of scrolling through hundreds of entries to see if you have enough technical documentation, you just ask your agent for the count. It turns a messy list of references into a structured, categorized dataset in seconds.

## Tools

### filter_sources_by_type
This tool isolates specific types of sources from your list. Use it to pull out only the legal or academic entries you need.

### get_source_counts
This tool calculates how many sources fall into each category. It gives you a direct tally of Academic, Media, or Technical types.

### get_total_source_summary
This tool provides a high level overview of your entire collection. It shows the total number of unique sources and how they are distributed.

### validate_source_schema
This tool checks your source data for structural errors. It ensures your list follows the required format before you start analyzing it.

## Prompt Examples

**Prompt:** 
```
How many sources do I have for each category in this list: [{"id": "1", "type": "Academic", "title": "Study A"}, {"id": "2", "type": "Media", "title": "News B"}]?
```

**Response:** 
```
You have 1 Academic source and 1 Media source.
```

**Prompt:** 
```
Give me a summary of these sources: [{"id": "1", "type": "Technical", "title": "Doc 1"}, {"id": "2", "type": "Technical", "title": "Doc 2"}]
```

**Response:** 
```
You have a total of 2 unique sources, with 2 belonging to the Technical category.
```

**Prompt:** 
```
Filter the sources to only show me the 'Legal' ones from this list: [{"id": "1", "type": "Legal", "title": "Patent X"}, {"id": "2", "type": "Academic", "title": "Paper Y"}]
```

**Response:** 
```
{"id": "1", "type": "Legal", "title": "Patent X"}
```

## Capabilities

### Data Categorization
Your agent uses this to group research materials into types like Media or Technical.

### Schema Validation
The AI checks your source lists to ensure they are formatted correctly.

### Source Distribution Analysis
Your agent calculates the ratio of different source types in a collection.

### Targeted Filtering
The AI extracts only the specific source types you ask for from a larger set.

## Use Cases

### Bibliography Auditing
Check if your research paper has enough academic sources compared to media sources.

### Data Cleaning
Verify that a list of research files matches the expected schema before processing.

### Competitive Intelligence
Categorize a large batch of news and technical documents to see where information is coming from.

### Literature Reviews
Get a quick tally of how many technical docs you have collected for a specific topic.

## Benefits

- Eliminates manual counting of research entries.
- Identifies gaps in source types immediately.
- Validates data integrity before analysis begins.
- Converts raw lists into structured summaries.

## How It Works

Connecting this MCP to your client gives your agent the ability to process research lists.

1. Connect the MCP to your client via Vinkius.
2. Provide a list of research sources to your AI client.
3. Ask your agent to count, filter, or summarize the data.
4. The MCP executes the specific tool requested.
5. Your agent delivers the formatted results.

## Frequently Asked Questions

**What can this MCP do with my research data?**
It can count sources by type, filter lists for specific categories, validate data schemas, and provide total summaries of your collections.

**Which AI clients can use this MCP?**
You can use this with any MCP-compatible client, including Claude, Cursor, Windsurf, and VS Code.

**How does it categorize sources?**
It uses the type field within your provided source data to group items into categories like Academic, Media, or Technical.

**Do I need to host this myself?**
No, Vinkius hosts and manages the MCP for you. You just connect it to your client and start using the tools.

**Can it check if my data is formatted correctly?**
Yes, the validate_source_schema tool specifically checks the schema of the sources you provide.

**How can I check if my source list is valid?**
You can use the `validate_source_schema` tool to verify that your list of sources contains the required unique identifiers and type classifications.

**Can I see a breakdown of my research types?**
Yes, the `get_source_counts` tool provides a specific count for every category present in your dataset.

**How do I get a total count of all my sources?**
Use the `get_total_source_summary` tool to receive both the total number of unique sources and a breakdown by type.