🤖 Salesforce MCP助理
使用MCP、FastAPI和Streamlit的端到端Salesforce自动化
______________________________________________________________________
🚀 概述
Salesforce MCP助理 是一款智能全栈人工智能助手，可使用 模型上下文协议（MCP）.\ 它允许用户通过一个干净的界面使用自然语言查询、创建、更新和管理Salesforce记录 流线型UI，由坚固的 快速API 后端。
该系统遵循模块化和可扩展的架构，将前端、后端和Salesforce MCP逻辑分开。\ 它的设计目的是 LLM就绪，使其易于与LangChain、OpenAI或其他人工智能框架集成，以实现高级自动化和推理。
______________________________________________________________________
✨ 主要特点
- 🤖 与Salesforce的自然语言交互
- 🔗 通过模型上下文协议（MCP）集成Salesforce
- ⚡ FastAPI后端用于可扩展的API处理
- 🖥️ 基于Streamlit的交互式前端
- 🧠 LLM就绪架构（兼容LangChain/OpenAI）
- 🧩 模块化和可扩展的项目结构
- 🔐 基于安全环境的凭据管理
______________________________________________________________________
🧱 项目结构
销售人员_MCP_助理/ │ ├── 后端/ │ ├── main.py#FastAPI服务器 │ ├── mcp_salesforce.py#用于处理查询的mcp包装器 │ └── salesforce_client.py#salesforce MCP客户端实现 │ ├── 前端/ │ └── app.py#基于流媒体的前端 │ ├── 需求.txt Mabalin遇到了i-print学生。 ├── README.md └── .env#出于安全考虑而忽略
YAML 复制代码
______________________________________________________________________
⚙️ 安装说明
1.️⃣ 克隆存储库
git clone https://github.com/Kartik-324/Salesforce_MCP_Assistant.git
cd Salesforce_MCP_Assistant
2️⃣ Create & Activate Virtual Environment
bash
Copy code
python -m venv venv
Windows
bash
Copy code
venv\Scripts\activate
macOS / Linux
bash
Copy code
source venv/bin/activate
3️⃣ Install Dependencies
bash
Copy code
pip install -r requirements.txt
▶️ Running the Application
🧠 Start Backend (FastAPI)
bash
Copy code
cd backend
uvicorn main:app --reload
💻 Start Frontend (Streamlit)
bash
Copy code
cd ../frontend
streamlit run app.py
Once both servers are running, open your browser at:
👉 http://localhost:8501
🔍 Example Queries
“Show me all Salesforce contacts”
“List accounts with revenue above 1M”
“Create an account named TechCorp”
“Get opportunities closed this month”
🧩 Tech Stack
Component Technology
Frontend Streamlit
Backend FastAPI
Integration Model Context Protocol (MCP)
Language Python
API Salesforce REST API
AI Ready LangChain / OpenAI Compatible
🛡️ Environment Variables
Create a .env file in the project root:
ini
Copy code
SALESFORCE_CLIENT_ID=your_client_id
SALESFORCE_CLIENT_SECRET=your_client_secret
SALESFORCE_USERNAME=your_username
SALESFORCE_PASSWORD=your_password
SALESFORCE_TOKEN=your_security_token
⚠️ Do not commit .env to GitHub
🧠 Powered By
Model Context Protocol (MCP)
FastAPI
Streamlit
LangChain (Optional Integration)
📌 Future Enhancements
🔄 Support for update & delete operations
🧠 Advanced LLM reasoning with LangGraph
📊 Analytics dashboard for Salesforce data
🔐 OAuth-based authentication
## 📄 License
This project is protected under a restrictive license.
The source code is shared for viewing and evaluation purposes only.
Any reuse or commercial use requires explicit permission.