deepseek-reasoning-mcp

Retrieve current, version-specific library documentation from the Context7 MCP server.

14|4|Updated Nov 22, 2025
One-click install
npx skills add https://github.com/pacphi/sindri --skill deepseek-reasoning-mcp
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: deepseek-reasoning-mcp
Source: https://github.com/pacphi/sindri/tree/main/docker/lib/extensions/vf-deepseek-reasoning/resources/mcp-server/server.js
Command: npx skills add https://github.com/pacphi/sindri --skill deepseek-reasoning-mcp

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill provides a server bridge to Direct DeepSeek reasoning endpoints for Claude.

Core Features & Use Cases

  • Reasoning Tools: deepseek_reason, deepseek_plan, deepseek_analyze.
  • SSH/Sudo Access: Executed as restricted user for credential isolation.

Quick Start

Connect and issue a reasoning task to DeepSeek.

Frequently Asked Questions about deepseek-reasoning-mcp

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I use DeepSeek reasoning capabilities with Claude?

DeepSeek reasoning tools—deepseek_reason, deepseek_plan, and deepseek_analyze—integrate directly with Claude through this MCP bridge. Connect the server and issue reasoning tasks to leverage DeepSeek's endpoints without managing separate API infrastructure.

Can I run DeepSeek reasoning tasks with restricted user access?

Yes, this Skill executes reasoning operations as a restricted user, isolating credentials and SSH/sudo access. This containment model prevents privilege escalation and protects API keys during Claude-to-DeepSeek communication.

What reasoning operations does the DeepSeek MCP bridge support?

The bridge exposes three core reasoning tools: deepseek_reason for general problem-solving, deepseek_plan for structured task decomposition, and deepseek_analyze for detailed examination. Each integrates with Claude's prompt context for collaborative reasoning workflows.

How does MCP enable DeepSeek integration without custom API code?

The Model Context Protocol standardizes server-to-LLM communication, allowing Claude to invoke DeepSeek reasoning endpoints through predefined tool schemas. This eliminates manual API wrappers and version management.

What's the difference between using DeepSeek directly versus through this MCP bridge?

Direct DeepSeek calls require separate API management and context switching. The MCP bridge embeds DeepSeek tools into Claude's native reasoning workflow, reducing latency, maintaining session context, and simplifying credential handling through restricted-user execution.