Neura Market
OthmaneBlial/term_mcp_deepseek logo

OthmaneBlial/term_mcp_deepseek

Free

用于终端的 DeepSeek 类 MCP 服务器。

FreeFree tier
Inputs: textOutputs: text
Type
Open Source
Visit Website GitHub (40 stars)

About OthmaneBlial/term_mcp_deepseek

An open-source, MCP-like server that leverages the DeepSeek API to enable AI-assisted terminal command execution. This proof-of-concept demonstrates core Model Context Protocol (MCP) concepts by exposing endpoints for tool listing and command invocation on an active shell session. It features a modern web chat interface (Flask + Tailwind CSS), real-time streaming via Server-Sent Events, persistent Bash session management, basic security measures (authentication, rate limiting), multiple transport options (HTTP REST API and STDIO), and Docker support for easy deployment. The server can interpret special CMD: instructions from the AI to trigger command execution. Note: This is not a fully compliant MCP server; additional work is needed for production readiness.

Key Features

Chat Interface: modern web-based chat client using Flask and Tailwind CSS with real-time updates
AI Integration: uses DeepSeek API to generate responses, can trigger commands via CMD: lines
Terminal Command Execution: executes shell commands via persistent Bash session using pexpect
MCP Endpoints: provides /mcp/list_tools and /mcp/call_tool endpoints mimicking MCP discovery and invocation
Real-time Streaming: Server-Sent Events (SSE) for live command execution updates
Security Enhancements: basic authentication, rate limiting, input validation, security headers
Multiple Transport Options: supports both HTTP REST API and STDIO command-line interface
Session Management: improved session handling and conversation storage
Docker Support: containerization for easier deployment and testing

Pros & Cons

Pros
  • Real-time streaming of command output via SSE
  • Session management with conversation storage
  • Multiple transport options (HTTP and STDIO) for flexibility
  • Docker support for easy deployment
  • Includes basic security features (authentication, rate limiting)
  • Open-source with permissive license
Cons
  • Not a fully compliant MCP server as defined by Anthropic
  • Requires a valid DeepSeek API key
  • Designed as a proof-of-concept, not production-ready
  • Limited to terminal command execution; no support for other MCP features like advanced authentication or comprehensive error handling

Best For

AI-assisted terminal command execution via natural languagePrototyping and learning Model Context Protocol (MCP) conceptsIntegrating DeepSeek AI with shell environments for automationEducational demonstrations of MCP server architecture