🚀Cursor降低智商！WindSurf零代码开发MCP Server！五分钟轻松实现LightRAG+MCP为Claude和AutoGen挂载知识库！

win4r2025-03-23 19:00

🚀🚀🚀本篇笔记所对应的视频：www.bilibili.com/video/BV1P6...

随着人工智能技术的不断突破，编程工具正迎来前所未有的变革。由 Codeium 团队推出的 WindSurf，以全新的 AI Flow 范式和多工具协同能力，正逐步超越备受关注的 Cursor，成为开发者提高工作效率的利器。

WindSurf 的最大亮点在于其深度上下文理解能力。传统的编程助手往往只能对简单代码片段进行补全，而 WindSurf 则能智能捕捉项目整体结构、变量关系以及函数调用链，无需开发者反复输入提示。它能够主动预测需求，在代码编写、重构、调试等过程中提供精准建议，从而大幅减少手动调试和反复确认的时间。

此外，WindSurf 内置了一整套工具集成系统。无论是文件搜索、目录管理，还是命令行执行，所有功能均实现无缝衔接。借助这一系统，开发者只需在一个平台内就能完成项目管理、代码编辑、依赖安装等多项任务。例如，在项目重构过程中，WindSurf 可以自动检测源目录与目标目录的状态，智能判断文件冲突，并根据历史操作自动选择最合适的执行策略，让整个过程变得流畅而高效。

与 Cursor 相比，WindSurf 在多步骤任务规划和协同工作方面优势更为明显。其支持多工具联动，通过自动维护上下文状态，能将复杂任务分解为多个子任务，逐步完成操作。这种能力不仅适用于大型项目的代码重构，也能帮助初学者在熟悉项目结构的同时，快速掌握开发技巧。

WindSurf 的智能编程助手功能更是为开发者提供了全新的交互体验。利用先进的 GPT-4o 和 Claude 3.5 模型，WindSurf 可处理文本、图像等多模态输入，生成高质量代码和解决方案。不仅如此，它还能根据开发者的操作习惯进行自动学习，将常用操作记录为隐式记忆，从而在后续对话中提供更加个性化的建议。

在实际应用中，比如一个基于 Nuxt 3 的 AI 工具集项目，开发者需要将旧有目录迁移至新架构。传统方法往往需要多次手动操作，而 WindSurf 则能自动扫描项目结构、识别配置文件、处理依赖问题，并利用内置命令顺利完成文件复制与删除、配置更新等任务。整个流程几乎无需人工干预，大大提升了项目重构效率。

未来，随着 AI 技术的进一步发展，WindSurf 将持续优化其智能交互和自动化能力，拓展更多企业级功能，如 SSO 认证、深度日志审计等。凭借强大的上下文感知、多步骤任务协同和灵活的工具集成，WindSurf 不仅重新定义了 AI 辅助编程的标准，更为开发者开辟了一条高效、智能的工作之路。

WindSurf 作为一款集智能、高效、便捷于一体的全新 AI 编程工具，正在引领编程工具的未来。无论是对个人开发者还是团队协作而言，WindSurf 都展示了它在提升代码质量、优化开发流程以及减少重复劳动方面的巨大潜力。未来的编程世界，将由这些智能工具驱动，让开发者真正实现"写代码，享受生活"。

WindSurf Prompt

sql 复制代码 
execute @task.md

虚拟环境创建

bash 复制代码 
python3 -m venv .venv
source .venv/bin/activate

Claude desktop app prompt

sql 复制代码 
 Using lightrag-server naive_search Tool to Search for "数据质量控制流程"，并用中文回答

AutoGen

ini 复制代码 
pip install -U "autogen-ext[mcp]"
pip install -U "autogen-agentchat"

import asyncio

from autogen_agentchat.agents import AssistantAgent
from autogen_ext.models.openai import OpenAIChatCompletionClient
from autogen_ext.tools.mcp import StdioServerParams, mcp_server_tools

async def main() -> None:
    # Get the fetch tool from mcp-server-fetch.
    rag_mcp_server = StdioServerParams(
        command="/Users/charlesqin/PycharmProjects/ligtrag-test/.venv/bin/python",
        args=["/Users/charlesqin/PycharmProjects/ligtrag-test/mcp_server.py"],
        env={"OPENAI_API_KEY": "sk-proj-xxxxxxxx"}
    )
    tools = await mcp_server_tools(rag_mcp_server)

    # Create an agent that can use the fetch tool.
    model_client = OpenAIChatCompletionClient(model="gpt-4o-mini")
    agent = AssistantAgent(name="fetcher", model_client=model_client, tools=tools, reflect_on_tool_use=True)  # type: ignore

    # Let the agent fetch the content of a URL and summarize it.
    result = await agent.run(task="Using lightrag-server naive_search Tool to Search for '数据质量控制流程'，并用中文回答")
    print(result.messages[-1].content)

asyncio.run(main())

MCP配置文件

json 复制代码 
{
  "mcpServers": {
    "lightrag": {
      "command": "/Users/charlesqin/PycharmProjects/ligtrag-test/.venv/bin/python",
      "args": ["/Users/charlesqin/PycharmProjects/ligtrag-test/mcp_server.py"],
      "env": {
        "OPENAI_API_KEY": "sk-proj---"
      }
    }
  }
}
ini 复制代码 
#--------------------------安装LightRAG--------------------------#

git clone <https://github.com/HKUDS/LightRAG.git>
cd LightRAG
pip install -e .
cd ..
export OPENAI_API_KEY=sk-proj-xxxxxxx

curl <https://raw.githubusercontent.com/win4r/mytest/refs/heads/main/book.txt> > ./book.txt

#--------------------------OpenAI--------------------------#

import os
import asyncio
from lightrag import LightRAG, QueryParam
from lightrag.llm.openai import gpt_4o_mini_complete, openai_embed
from lightrag.kg.shared_storage import initialize_pipeline_status

WORKING_DIR = "./mybook"

if not os.path.exists(WORKING_DIR):
    os.mkdir(WORKING_DIR)

async def initialize_rag():
    rag = LightRAG(
        working_dir=WORKING_DIR,
        embedding_func=openai_embed,
        llm_model_func=gpt_4o_mini_complete,
        # llm_model_func=gpt_4o_complete
    )

    await rag.initialize_storages()
    await initialize_pipeline_status()

    return rag

def main():
    # Initialize RAG instance
    rag = asyncio.run(initialize_rag())

    with open("./book.txt", "r", encoding="utf-8") as f:
        rag.insert(f.read())

    # Perform naive search
    print(
        rag.query(
            "What is the Self-Consistency Prompt?", param=QueryParam(mode="naive")
        )
    )

    # Perform local search
    print(
        rag.query(
            "What is the Self-Consistency Prompt?", param=QueryParam(mode="local")
        )
    )

    # Perform global search
    print(
        rag.query(
            "What is the Self-Consistency Prompt?", param=QueryParam(mode="global")
        )
    )

    # Perform hybrid search
    print(
        rag.query(
            "What are the top themes in this story?", param=QueryParam(mode="hybrid")
        )
    )

if __name__ == "__main__":
    main()
    

#--------------------------ollama--------------------------#

  
import asyncio
import nest_asyncio

nest_asyncio.apply()
import os
import inspect
import logging
from lightrag import LightRAG, QueryParam
from lightrag.llm.ollama import ollama_model_complete, ollama_embed
from lightrag.utils import EmbeddingFunc
from lightrag.kg.shared_storage import initialize_pipeline_status

WORKING_DIR = "./dickens"

logging.basicConfig(format="%(levelname)s:%(message)s", level=logging.INFO)

if not os.path.exists(WORKING_DIR):
    os.mkdir(WORKING_DIR)

async def initialize_rag():
    rag = LightRAG(
        working_dir=WORKING_DIR,
        llm_model_func=ollama_model_complete,
        llm_model_name="gemma2:2b",
        llm_model_max_async=4,
        llm_model_max_token_size=32768,
        llm_model_kwargs={
            "host": "<http://localhost:11434>",
            "options": {"num_ctx": 32768},
        },
        embedding_func=EmbeddingFunc(
            embedding_dim=768,
            max_token_size=8192,
            func=lambda texts: ollama_embed(
                texts, embed_model="nomic-embed-text", host="<http://localhost:11434>"
            ),
        ),
    )

    await rag.initialize_storages()
    await initialize_pipeline_status()

    return rag

async def print_stream(stream):
    async for chunk in stream:
        print(chunk, end="", flush=True)

def main():
    # Initialize RAG instance
    rag = asyncio.run(initialize_rag())

    # Insert example text
    with open("./book.txt", "r", encoding="utf-8") as f:
        rag.insert(f.read())

    # Test different query modes
    print("\nNaive Search:")
    print(
        rag.query(
            "What are the top themes in this story?", param=QueryParam(mode="naive")
        )
    )

    print("\nLocal Search:")
    print(
        rag.query(
            "What are the top themes in this story?", param=QueryParam(mode="local")
        )
    )

    print("\nGlobal Search:")
    print(
        rag.query(
            "What are the top themes in this story?", param=QueryParam(mode="global")
        )
    )

    print("\nHybrid Search:")
    print(
        rag.query(
            "What are the top themes in this story?", param=QueryParam(mode="hybrid")
        )
    )

    # stream response
    resp = rag.query(
        "What are the top themes in this story?",
        param=QueryParam(mode="hybrid", stream=True),
    )

    if inspect.isasyncgen(resp):
        asyncio.run(print_stream(resp))
    else:
        print(resp)

if __name__ == "__main__":
    main()

windsurf Prompt

vbnet 复制代码 
### task:

Build an MCP server for the LightRAG.
This server should allow me to input any query text and perform search operations using four distinct MCP Tools: naive search, local search, global search, and hybrid search.

Support Multiple Query Modes:

Naive Search: Perform a straightforward search over the inserted content.

Local Search: Focus on retrieving results within a limited, local context of the inserted documents.

Global Search: Conduct a comprehensive search across the entire dataset.

Hybrid Search: Combine aspects of local and global searches to enhance result relevance and consistency.

The WORKING_DIR path is /Users/charlesqin/PycharmProjects/ligtrag-test

Please activate the venv:
cd /Users/charlesqin/PycharmProjects/ligtrag-test
source .venv/bin/activate

Here's the Code and Response for the LightRAG:

### Code (app.py):

import os
import asyncio
from lightrag import LightRAG, QueryParam
from lightrag.llm.openai import gpt_4o_mini_complete, openai_embed
from lightrag.kg.shared_storage import initialize_pipeline_status

WORKING_DIR = "./mybook"

if not os.path.exists(WORKING_DIR):
    os.mkdir(WORKING_DIR)

async def initialize_rag():
    rag = LightRAG(
        working_dir=WORKING_DIR,
        embedding_func=openai_embed,
        llm_model_func=gpt_4o_mini_complete,
        # llm_model_func=gpt_4o_complete
    )

    await rag.initialize_storages()
    await initialize_pipeline_status()

    return rag

def main():
    # Initialize RAG instance
    rag = asyncio.run(initialize_rag())

    with open("./book.txt", "r", encoding="utf-8") as f:
        rag.insert(f.read())

    # Perform naive search
    print(
        rag.query(
            "What is the Self-Consistency Prompt?", param=QueryParam(mode="naive")
        )
    )

    # Perform local search
    print(
        rag.query(
            "What is the Self-Consistency Prompt?", param=QueryParam(mode="local")
        )
    )

    # Perform global search
    print(
        rag.query(
            "What is the Self-Consistency Prompt?", param=QueryParam(mode="global")
        )
    )

    # Perform hybrid search
    print(
        rag.query(
            "What are the top themes in this story?", param=QueryParam(mode="hybrid")
        )
    )

if __name__ == "__main__":
    main()
    
    

### Response:

[Full response logs showing the initialization and search results]

(.venv) charlesqin@charless-MacBook-Pro ligtrag-test % python app.py  
INFO: Process 31212 Shared-Data created for Single Process
INFO:nano-vectordb:Load (89, 1536) data
INFO:nano-vectordb:Init {'embedding_dim': 1536, 'metric': 'cosine', 'storage_file': './mybook/vdb_entities.json'} 89 data
INFO:nano-vectordb:Load (85, 1536) data
INFO:nano-vectordb:Init {'embedding_dim': 1536, 'metric': 'cosine', 'storage_file': './mybook/vdb_relationships.json'} 85 data
INFO:nano-vectordb:Load (9, 1536) data
INFO:nano-vectordb:Init {'embedding_dim': 1536, 'metric': 'cosine', 'storage_file': './mybook/vdb_chunks.json'} 9 data
INFO: Process 31212 initialized updated flags for namespace: [full_docs]
INFO: Process 31212 ready to initialize storage namespace: [full_docs]
INFO: Process 31212 initialized updated flags for namespace: [text_chunks]
INFO: Process 31212 ready to initialize storage namespace: [text_chunks]
INFO: Process 31212 initialized updated flags for namespace: [entities]
INFO: Process 31212 initialized updated flags for namespace: [relationships]
INFO: Process 31212 initialized updated flags for namespace: [chunks]
INFO: Process 31212 initialized updated flags for namespace: [chunk_entity_relation]
INFO: Process 31212 initialized updated flags for namespace: [llm_response_cache]
INFO: Process 31212 ready to initialize storage namespace: [llm_response_cache]
INFO: Process 31212 initialized updated flags for namespace: [doc_status]
INFO: Process 31212 ready to initialize storage namespace: [doc_status]
INFO: Process 31212 storage namespace already initialized: [full_docs]
INFO: Process 31212 storage namespace already initialized: [text_chunks]
INFO: Process 31212 storage namespace already initialized: [llm_response_cache]
INFO: Process 31212 storage namespace already initialized: [doc_status]
INFO: Process 31212 Pipeline namespace initialized
## Self-Consistency Prompt

The Self-Consistency prompt is a technique utilized to ensure that the output generated by ChatGPT maintains consistency with the provided input. This approach is particularly useful for tasks such as fact-checking, data validation, or ensuring consistency in text generation.

### Prompt Formula
The prompt formula for the Self-Consistency prompt involves embedding the input text alongside an instruction. The instruction typically states, "Please ensure the following text is self-consistent."

### Examples of Application

1. **Text Generation**
   - **Task**: Generate a product review.
   - **Instructions**: The review should be consistent with the product information provided in the input.
   - **Prompt formula**: "Generate a product review that is consistent with the following product information [insert product information]."

2. **Text Summarization**
   - **Task**: Summarize a news article.
   - **Instructions**: The summary should be consistent with the information provided in the article.
   - **Prompt formula**: "Summarize the following news article in a way that is consistent with the information provided [insert news article]."

3. **Fact-checking**
   - **Task**: Check for consistency in a given news article.
   - **Input text**: "The article states that the population of the city is 5 million, but later on, it says that the population is 7 million."
   - **Prompt formula**: "Please ensure the following text is self-consistent: The article states that the population of the city is 5 million, but later on, it says that the population is 7 million."

4. **Data Validation**
   - **Task**: Check for consistency in a given data set.
   - **Input text**: "The data shows that the average temperature in July is 30 degrees, but the minimum temperature is recorded as 20 degrees."
   - **Prompt formula**: "Please ensure the following text is self-consistent: The data shows that the average temperature in July is 30 degrees, but the minimum temperature is recorded as 20 degrees."

By employing the Self-Consistency prompt, users can enhance the reliability of the text generated by ensuring that it aligns with the provided inputs.

### References
- [KG] Self-Consistency Prompt Description (File: Document Chunks)
- [KG] Application Examples for Self-Consistency Prompts (File: Document Chunks)
The Knowledge Base does not provide specific information on the "Self-Consistency Prompt." It typically discusses various prompt engineering techniques and their applications with language models like ChatGPT, but mentions self-consistency without elaborating on its definition or use cases.

If you have more detailed questions about other prompt techniques or topics related to prompt engineering, please feel free to ask!

### References
- [KG] Prompt Engineering Techniques encompass a variety of methods and practices for effectively engaging with language models like ChatGPT. (File: unknown_source)
- [KG] The Art of Asking ChatGPT for High-Quality Answers: A Complete Guide to Prompt Engineering Techniques covers various techniques related to prompt engineering and the effective use of ChatGPT. (File: unknown_source)
- [KG] ChatGPT is a state-of-the-art language model capable of generating human-like text based on the prompts given. (File: unknown_source)
The Knowledge Base provided does not contain specific information about the "Self-Consistency Prompt." This term may refer to a prompting technique meant to encourage models, like ChatGPT, to generate consistent outputs across different prompts or contexts, but further details are not available in the provided content. Therefore, I cannot provide a detailed explanation or context for the Self-Consistency Prompt.

If you have any other questions or need information on a different topic, feel free to ask! 

### References
- [KG] Ibrahim John discusses various prompting techniques in his book (File: unknown_source)
- [KG] Knowledge generation techniques are outlined in the book's content (File: unknown_source)
I don't have specific details about the story you're referring to, as the Knowledge Base does not provide information on any particular narrative beyond discussing techniques for text generation and prompt engineering. 

However, common themes in literature often include:

1. **Love** - exploring emotional connections and relationships.
2. **Personal Growth** - focusing on the journey of self-discovery and development.
3. **Conflict** - presenting struggles between characters or within oneself.
4. **Change** - highlighting transformations in characters or settings (e.g., changing seasons).
5. **Morality** - examining the concepts of right and wrong.

If you could provide more details or specify the story, I could help further! 

### References
- [KG] Love is a complex emotion that forms the basis of many poems and literary works. (File: unknown_source)
- [KG] Personal growth refers to the process of self-improvement and development, focusing on various aspects of life such as emotional, intellectual and relational growth. (File: unknown_source)
- [KG] The changing seasons refer to the transition periods throughout the year characterized by different weather patterns, environmental changes, and their impact on nature and human life. (File: unknown_source)
(.venv) charlesqin@charless-MacBook-Pro ligtrag-test % 

### The MCP document is:
⚠️请将下面链接里的MCP文档完整的复制粘贴到这里，由于博客字数有限，所以请自行复制文档内容⬇️
<https://modelcontextprotocol.io/llms-full.txt>
相关推荐
AI绘画小3332 分钟前
Stable Diffusion教程：图生图_stable diffusion 图生图
人工智能·ai作画·stable diffusion·aigc·图生图·ai教程·sd教程
NineData4 小时前
9行代码实现MCP Server示例
人工智能·openai·mcp
PaperTen4 小时前
ChatGPT降低论文AIGC重复率的提示词合集（高效降重方法）
chatgpt·aigc
加勒比大风车4 小时前
Suno AI使用教程：从入门到进阶(超详细)
aigc·openai
yaocheng的ai分身5 小时前
cursor又更新大版本了，比牛还勤快
cursor·mcp
牛奶9 小时前
浅见：在AI浪潮中的思考与前行
前端·人工智能·openai
九歌AI大模型9 小时前
Dify多版本Windows虚拟机，10分钟傻瓜式部署智能体开发平台！
aigc
Jokeny9 小时前
百度秒哒上线，AI无代码平台大战一触即发，谁将笑到最后？
aigc
运营黑客9 小时前
Deepseek-v3重磅更新，能力直追Claude-3.7：一句话生成精美卡片、PPT、动图、可视化网站
aigc·deepseek
abo1234569 小时前
LHM：单张图片生成 3D 可动画人物的AI革新技术
aigc
热门推荐
01LLaMA-Factory DeepSeek-R1 模型 微调基础教程02RAG 实践- Ollama+RagFlow 部署本地知识库03DeepSeek各版本说明与优缺点分析04本地化部署AI知识库：基于Ollama+DeepSeek+AnythingLLM保姆级教程05如何在WPS和Word/Excel中直接使用DeepSeek功能06从零安装 LLaMA-Factory 微调 Qwen 大模型成功及所有的坑07FreeCAD傻瓜教程之创建参数化几何图形-螺旋体、平面、球体、椭球体、圆柱体、圆锥体、棱柱、椭圆08本地部署DeepSeek教程（Mac版本）09DeepSeek R1本地化部署 Ollama + Chatbox 打造最强 AI 工具10windows通过FVM管理多个版本flutter