gptr-mcp

概述

GPT Researcher MCP 是一个基于Python的服务器，通过MCP协议让LLM应用执行深度研究。与返回原始结果的标准搜索工具不同，该服务器自主探索和验证来源，提供高质量、全面的研究成果。它既提供快速搜索以满足速度需求，也提供深度研究以确保全面性，同时还具备报告生成功能。服务器支持多种传输协议，包括本地开发的STDIO和Docker/部署的SSE。

试试问 AI

装完之后，这里有 7 个你可以让 AI 做的事：

• research_resource

  Get web resources related to a given task via research.

  试试:Find web resources about renewable energy investments试试:Get resources related to AI ethics in healthcare
• deep_research

  Performs deep web research on a topic, finding the most reliable and relevant information.

  试试:Research the latest developments in quantum computing试试:Investigate climate change impacts on coastal cities
• quick_search

  Performs a fast web search optimized for speed over quality.

  试试:Quick search for recent tech news试试:Fast search for Python tutorials
• write_report

  Generate a report based on research results.

  试试:Write a report summarizing the research findings试试:Generate a comprehensive report on market trends
• get_research_sources

  Get the sources used in the research.

  试试:Show me all the sources used for this research试试:Get the research references for my query
• get_research_context

  Get the full context of the research.

  试试:Provide the full research context试试:Show me all the research material
• research_query

  Create a research query prompt.

  试试:Help me formulate a research query about blockchain technology试试:Create a query to research stock market trends

可对比工具

安装

git clone https://github.com/assafelovic/gpt-researcher.git
cd gpt-researcher/cd gptr-mcp

pip install -r requirements.txt

cp .env.example .env
# 编辑.env文件添加您的API密钥

{
  "mcpServers": {
    "gptr-mcp": {
      "command": "python",
      "args": ["/绝对路径/to/gptr-mcp/server.py"],
      "env": {
        "OPENAI_API_KEY": "您的-openai-密钥",
        "TAVILY_API_KEY": "您的-tavily-密钥"
      }
    }
  }
}

FAQ

深度研究和快速搜索有什么区别？

深度研究自主探索并验证多个来源，提供全面高质量的信息（耗时约30秒）。快速搜索则是为速度优化的快速网络搜索，返回带有片段的基本搜索结果。

需要哪些API密钥？

您需要一个OpenAI API密钥用于LLM，以及一个Tavily API密钥用于网络搜索。如果需要，可以配置其他搜索引擎。

gptr-mcp 对比