# Deepseek

> Use this tool when you need to extract and utilize step-by-step reasoning chains for complex problem-solving, decision analysis, and explainable AI. Deepseek provides a bridge between its API and MCP-compatible applications, allowing developers and researchers to incorporate advanced reasoning into their workflows. It takes in requests from MCP clients and outputs reasoning chains, enabling standardized access to Deepseek R1's capabilities.

Canonical page: https://skillsregistry.net/skills/matt-carey-deepseek-reasoning  
JSON: https://api.skillsregistry.net/v1/skills/matt-carey-deepseek-reasoning

## Description

This MCP thinking server, developed by Matt Carey, extracts the reasoning chain from the Deepseek R1 model and makes it available to Claude Desktop and other MCP clients. Built with TypeScript and leveraging the Model Context Protocol SDK, it provides a bridge between Deepseek's API and MCP-compatible applications. The implementation focuses on exposing Deepseek R1's step-by-step reasoning capabilities through a standardized interface. It's particularly useful for developers and researchers looking to incorporate Deepseek's advanced reasoning into their workflows, enabling use cases like complex problem-solving, decision analysis, and explainable AI without directly interacting with Deepseek's API.

## Trust

- **Trust score (0–1):** 0.95
- **Verification tier:** verified
- **Last scanned:** 2026-09-19

## Facts

- **Version:** 1.0.0
- **Skill type:** atomic
- **Execution layer:** mcp-remote
- **Runtime environment:** api
- **Category:** search
- **Updated:** 2026-09-19

## Source

- **Source listing:** [PulseMCP](https://www.pulsemcp.com/servers/matt-carey-deepseek-reasoning)
- **Repository:** <https://github.com/mattzcarey/mcp-thinking>

## Use it

Resolve this record through the SkillsRegistry MCP server (no auth, read-only):

```
claude mcp add --transport http --scope user skillsregistry https://api.skillsregistry.net/mcp
```

```json
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "get_skill",
    "arguments": {
      "slug": "matt-carey-deepseek-reasoning"
    }
  }
}
```

REST: `GET https://api.skillsregistry.net/v1/skills/matt-carey-deepseek-reasoning` · pull for local use: `GET https://api.skillsregistry.net/v1/skills/matt-carey-deepseek-reasoning/pull`

---
SkillsRegistry indexes agent skills from public registries and GitHub. Skills we have analysed are scanned with Circle-IR and scored on six dimensions; each listing states its scan coverage. More: https://skillsregistry.net/llms.txt