diff --git a/ci/tools/DocsRAG_test.py b/ci/tools/DocsRAG_test.py
new file mode 100644
index 0000000000000000000000000000000000000000..c7c19ab3b20543e4617559f35b03ea1ddc223808
--- /dev/null
+++ b/ci/tools/DocsRAG_test.py
@@ -0,0 +1,92 @@
+import unittest
+from unittest.mock import Mock, patch
+from langchain.embeddings.base import Embeddings
+from langchain_community.vectorstores import FAISS
+from myFirstRagDemo import (
+    load_knowledge_base,
+    create_vector_db,
+    setup_rag_chain,
+    get_local_embeddings,
+    get_deepseek_llm
+)
+import os
+
+class TestRAGDemo(unittest.TestCase):
+    def test_load_knowledge_base(self):
+        """测试知识库加载功能"""
+        # 执行函数
+        knowledge_base = load_knowledge_base()
+        
+        # 验证结果
+        self.assertEqual(knowledge_base.strip(), "This is a test text.")
+        self.assertIsInstance(knowledge_base, str)
+    
+    def test_create_vector_db(self):
+        """测试向量数据库创建功能"""
+        # 创建模拟嵌入模型
+        class MockEmbeddings(Embeddings):
+            def embed_query(self, text):
+                return [0.1] * 768  # 假设BERT输出768维向量
+            
+            def embed_documents(self, texts):
+                return [[0.1] * 768 for _ in texts]
+        
+        mock_embeddings = MockEmbeddings()
+        test_text = "这是一个测试文本，用于验证向量数据库创建功能。"
+        
+        # 执行函数
+        vector_db = create_vector_db(test_text, mock_embeddings)
+        
+        # 验证结果
+        self.assertIsInstance(vector_db, FAISS)
+        self.assertEqual(len(vector_db.docstore._dict), 1)  # 检查文档数量
+    
+    def test_setup_rag_chain(self):
+        """测试RAG链设置功能"""
+        # 创建模拟向量数据库
+        class MockFAISS(FAISS):
+            def as_retriever(self, search_kwargs=None):
+                return Mock()  # 返回模拟检索器
+        
+        mock_vector_db = MockFAISS()
+        mock_llm = Mock()  # 模拟语言模型
+        
+        # 执行函数
+        rag_chain = setup_rag_chain(mock_vector_db, mock_llm)
+        
+        # 验证结果
+        self.assertIsNotNone(rag_chain)
+        self.assertEqual(rag_chain.return_source_documents, True)
+    
+    def test_get_local_embeddings(self):
+        """测试本地嵌入模型加载功能"""
+        # 执行函数
+        embeddings = get_local_embeddings()
+        
+        # 验证结果
+        self.assertIsNotNone(embeddings)
+        self.assertEqual(embeddings.model_name, 'bert-base-chinese')
+        self.assertEqual(embeddings.model_kwargs, {"device": "cpu"})
+    
+    @patch('langchain_deepseek.ChatDeepSeek')
+    def test_get_deepseek_llm(self, mock_chat_deepseek):
+        """测试DeepSeek语言模型加载功能"""
+        # 设置测试API密钥
+        test_api_key = "test_api_key_123"
+        
+        # 执行函数
+        llm = get_deepseek_llm(test_api_key)
+        
+        # 验证环境变量是否正确设置
+        self.assertEqual(os.environ["DEEPSEEK_API_KEY"], test_api_key)
+        
+        # 验证ChatDeepSeek是否被正确调用
+        mock_chat_deepseek.assert_called_once_with(
+            model="deepseek-chat",
+            temperature=0,
+            max_tokens=1024
+        )
+        self.assertIsInstance(llm, Mock)  # 由于使用了mock，应该返回一个Mock对象
+
+if __name__ == '__main__':
+    unittest.main()
\ No newline at end of file
diff --git a/ci/tools/DocsRag.py b/ci/tools/DocsRag.py
new file mode 100644
index 0000000000000000000000000000000000000000..5bad541bd3fa79ca0b29b796c6f1091255947774
--- /dev/null
+++ b/ci/tools/DocsRag.py
@@ -0,0 +1,117 @@
+# To run this script, we need pip some component like this:.
+#    1 pip install langchain
+#    2 python.exe -m pip install --upgrade pip
+#    3 pip install langchain-community
+#    4 pip install langchain_deepseek
+#    5 pip install sentence-transformers
+#    6 pip install -U langchain-huggingface
+#    7 pip install hf_xet
+#    8 pip install faiss-cpu
+import argparse
+from langchain.embeddings.base import Embeddings
+from langchain_community.vectorstores import FAISS
+from langchain.text_splitter import RecursiveCharacterTextSplitter
+from langchain_community.embeddings import HuggingFaceEmbeddings
+import os
+from langchain_deepseek import ChatDeepSeek
+from langchain.chains import RetrievalQA
+
+# Press Shift+F10 to execute it or replace it with your code.
+# Press Double Shift to search everywhere for classes, files, tool windows, actions, and settings.
+
+# LOCAL Knowledge
+def load_knowledge_base()->str:
+    knowledge_base = """
+This is a test text.
+"""
+    return knowledge_base
+
+#2.分开和向量化
+def create_vector_db(txt:str, embeddings: Embeddings) -> FAISS:
+    text_splitter = RecursiveCharacterTextSplitter(
+        chunk_size=100,#每块最多100字符
+        chunk_overlap=20,#相邻重叠20字符
+    )
+    # 生产适合大小文本
+    chunks = text_splitter.create_documents([txt])
+
+    # 创建向量
+    vector_db = FAISS.from_documents(chunks, embeddings)
+    return vector_db
+
+#3.检索生成
+def setup_rag_chain(vector_db:FAISS, llm) ->RetrievalQA:
+    #创检索器
+    retriever = vector_db.as_retriever(search_kwargs={"k": 2})
+    #创RAG链
+    rag_chain = RetrievalQA.from_chain_type(
+        llm=llm,
+        chain_type="stuff",
+        retriever=retriever,
+        return_source_documents=True
+    )
+    return rag_chain
+
+#获取embeddings
+def get_local_embeddings() -> HuggingFaceEmbeddings:
+    """加载本地嵌入模型，兼容TensorFlow格式"""
+
+    embedder = HuggingFaceEmbeddings(
+        model_name='bert-base-chinese',
+        model_kwargs={"device": "cpu"}
+    )
+    return embedder
+
+#获取语言模型
+def get_deepseek_llm(api_key:str):
+    os.environ["DEEPSEEK_API_KEY"] = api_key
+    return ChatDeepSeek(
+        model="deepseek-chat",  # 支持模型如deepseek-chat（DeepSeek-V3）、deepseek-reasoner（DeepSeek-R1）
+        temperature=0,
+        max_tokens=1024
+    )
+
+
+def main():
+    parser = argparse.ArgumentParser(description="First RAG demo.")
+    parser.add_argument("--api_key", type=str, default="*****", help="Please input AI API key.")
+
+    args = parser.parse_args()
+    key = args.api_key
+    if not key:
+        print("The Key input Error.")
+        return
+
+    #加载文档库
+    knowledge_base = load_knowledge_base()
+
+    #创建embedding模型
+    embeddings = get_local_embeddings()
+    vector_db = create_vector_db(knowledge_base, embeddings)
+
+    #创建llm
+    print(key)
+    llm = get_deepseek_llm(key)
+    rag_chain = setup_rag_chain(vector_db, llm)
+
+    print("\nRAG系统已就绪，请输入问题(输入'Q'结束对话):")
+    while True:
+        user_query = input("\n问题: ")
+        if user_query.lower() == "Q":
+            break
+
+        # 执行RAG流程
+        result = rag_chain({"query": user_query})
+
+        # 显示结果
+        print("\n回答:")
+        print(result["result"])
+
+        # 显示来源
+        print("\n参考内容:")
+        for doc in result["source_documents"]:
+            print(f"- {doc.page_content[:100]}...")
+
+# Press the green button in the gutter to run the script.
+if __name__ == '__main__':
+    main()