✨ feat(database_optimizer): 添加数据库优化工程师智能体功能
- 使用 DeepSeek 模型进行智能分析 - 集成 PostgreSQL MCP 服务器工具 - 加载数据库知识库 - 提供数据库优化建议
This commit is contained in:
125
数据库优化工程师.py
125
数据库优化工程师.py
@@ -1,96 +1,109 @@
|
||||
if __name__ == "__main__":
|
||||
"""
|
||||
## 背景
|
||||
#!/usr/bin/env python3
|
||||
"""
|
||||
PostgreSQL 数据库优化工程师智能体
|
||||
|
||||
我有一个`postgresql`数据库,现在需要一个专业的数据库优化工程师,来帮我看下设计是否合理,以及如何优化.
|
||||
功能:
|
||||
- 使用 DeepSeek 模型进行智能分析
|
||||
- 集成 PostgreSQL MCP 服务器工具
|
||||
- 加载数据库知识库
|
||||
- 提供数据库优化建议
|
||||
"""
|
||||
|
||||
## 目标
|
||||
import os
|
||||
from typing import Optional
|
||||
from dataclasses import dataclass
|
||||
from dotenv import load_dotenv
|
||||
|
||||
这个数据库优化工程师智能体应该具备以下能力:
|
||||
from agno.agent import Agent
|
||||
from agno.models.deepseek import DeepSeek
|
||||
from agno.knowledge.pdf import PDFKnowledgeBase
|
||||
from agno.vectordb.lancedb import LanceDb, SearchType
|
||||
from agno.embedder.openai import OpenAIEmbedder
|
||||
from agno.knowledge.combined import CombinedKnowledgeBase
|
||||
from agno.tools.mcp import MCPTools
|
||||
from mcp import StdioServerParameters
|
||||
|
||||
- [ ] 使用`claude3.7`作为模型(相当于拥有一个聪明的大脑,总是能够做出正确的决策)
|
||||
- [ ] 精通关系型数据库系统的理论知识以及`postgresql`的实现细节(知识库)
|
||||
- [ ] 能够使用`postgresql mcp server`来分析现有数据库设计(行为)
|
||||
@dataclass
|
||||
class DatabaseOptimizerConfig:
|
||||
"""数据库优化器配置"""
|
||||
pdf_path: str = "D:\\Sources\\DONGJAK-TOOLS\\pdfs\\Database Fundamentals.pdf"
|
||||
db_connection: str = "postgresql://postgres:postgres@192.168.1.7:5432/postgres"
|
||||
model_id: str = "deepseek-chat"
|
||||
vector_db_path: str = "tmp/lancedb"
|
||||
|
||||
"""
|
||||
# 然后在notebook中加载
|
||||
from dotenv import load_dotenv
|
||||
class DatabaseOptimizer:
|
||||
"""PostgreSQL 数据库优化引擎"""
|
||||
|
||||
# 加载当前目录下的.env文件
|
||||
def __init__(self, config: Optional[DatabaseOptimizerConfig] = None):
|
||||
self.config = config or DatabaseOptimizerConfig()
|
||||
self._load_environment()
|
||||
self.knowledge_base = self._setup_knowledge_base()
|
||||
self.postgres_tools = self._setup_postgres_tools()
|
||||
self.agent = self._create_agent()
|
||||
|
||||
def _load_environment(self):
|
||||
"""加载环境变量"""
|
||||
load_dotenv()
|
||||
from agno.agent import Agent, RunResponse # noqa
|
||||
from agno.models.deepseek import DeepSeek
|
||||
from agno.knowledge.pdf_url import PDFUrlKnowledgeBase
|
||||
from agno.knowledge.pdf import PDFKnowledgeBase
|
||||
from agno.vectordb.lancedb import LanceDb, SearchType
|
||||
from agno.embedder.openai import OpenAIEmbedder
|
||||
from agno.knowledge.combined import CombinedKnowledgeBase
|
||||
from agno.tools.mcp import MCPTools
|
||||
from mcp import StdioServerParameters
|
||||
|
||||
# Create a knowledge base of PDFs from URLs
|
||||
# pdf_url_kb = PDFUrlKnowledgeBase(
|
||||
# urls=["https://agno-public.s3.amazonaws.com/recipes/ThaiRecipes.pdf"],
|
||||
# # Use LanceDB as the vector database and store embeddings in the `recipes` table
|
||||
# vector_db=LanceDb(
|
||||
# table_name="recipes",
|
||||
# uri="tmp/lancedb",
|
||||
# search_type=SearchType.vector,
|
||||
# embedder=OpenAIEmbedder(id="text-embedding-3-small"),
|
||||
# ),
|
||||
# )
|
||||
# Create Local PDF knowledge base
|
||||
def _setup_knowledge_base(self) -> CombinedKnowledgeBase:
|
||||
"""设置知识库"""
|
||||
local_pdf_kb = PDFKnowledgeBase(
|
||||
path="D:\\Sources\\DONGJAK-TOOLS\\pdfs\\Database Fundamentals.pdf",
|
||||
path=self.config.pdf_path,
|
||||
vector_db=LanceDb(
|
||||
table_name="database_fundamentals",
|
||||
uri="tmp/lancedb",
|
||||
uri=self.config.vector_db_path,
|
||||
search_type=SearchType.vector,
|
||||
embedder=OpenAIEmbedder(id="text-embedding-3-small"),
|
||||
),
|
||||
)
|
||||
|
||||
# Combine knowledge bases
|
||||
knowledge_base = CombinedKnowledgeBase(
|
||||
sources=[
|
||||
local_pdf_kb,
|
||||
],
|
||||
sources=[local_pdf_kb],
|
||||
vector_db=LanceDb(
|
||||
table_name="combined_documents",
|
||||
uri="tmp/lancedb",
|
||||
uri=self.config.vector_db_path,
|
||||
search_type=SearchType.vector,
|
||||
embedder=OpenAIEmbedder(id="text-embedding-3-small"),
|
||||
),
|
||||
)
|
||||
# Load the knowledge base: Comment after first run as the knowledge base is already loaded
|
||||
knowledge_base.load()
|
||||
return knowledge_base
|
||||
|
||||
def _setup_postgres_tools(self) -> MCPTools:
|
||||
"""设置 PostgreSQL 工具"""
|
||||
server_params = StdioServerParameters(
|
||||
command="cmd", # 或 "uvx"，取决于你的安装方式
|
||||
command="cmd",
|
||||
args=[
|
||||
"/c",
|
||||
"npx",
|
||||
"-y",
|
||||
"@modelcontextprotocol/server-postgres",
|
||||
"postgresql://postgres:postgres@192.168.1.7:5432/postgres",
|
||||
self.config.db_connection,
|
||||
],
|
||||
env={}, # 可选的环境变量
|
||||
env={},
|
||||
)
|
||||
with MCPTools(server_params=server_params) as postgres_server:
|
||||
# 使用mcp_tools
|
||||
return MCPTools(server_params=server_params)
|
||||
|
||||
agent = Agent(
|
||||
model=DeepSeek(id="deepseek-chat"),
|
||||
def _create_agent(self) -> Agent:
|
||||
"""创建智能体"""
|
||||
return Agent(
|
||||
model=DeepSeek(id=self.config.model_id),
|
||||
markdown=True,
|
||||
knowledge=knowledge_base,
|
||||
knowledge=self.knowledge_base,
|
||||
search_knowledge=True,
|
||||
show_tool_calls=True,
|
||||
tools=[postgres_server],
|
||||
tools=[self.postgres_tools],
|
||||
)
|
||||
|
||||
# Get the response in a variable
|
||||
# run: RunResponse = agent.run("Share a 2 sentence horror story")
|
||||
# print(run.content)
|
||||
def analyze_database(self, query: str, stream: bool = True):
|
||||
"""分析数据库"""
|
||||
with self.postgres_tools:
|
||||
self.agent.print_response(query, stream=stream)
|
||||
|
||||
# Print the response in the terminal
|
||||
agent.print_response("看下aq这个数据库", stream=True)
|
||||
def main():
|
||||
"""主入口函数"""
|
||||
optimizer = DatabaseOptimizer()
|
||||
optimizer.analyze_database("看下aq这个数据库")
|
||||
|
||||
if __name__ == "__main__":
|
||||
main()
|
||||
|
||||
Reference in New Issue
Block a user