xgae 0.2.0__tar.gz → 0.2.1__tar.gz

{xgae-0.2.0 → xgae-0.2.1}/.env

@@ -12,6 +12,7 @@ LANGFUSE_HOST=https://cloud.langfuse.com
 
 # LLM
 LLM_MODEL=openai/qwen3-235b-a22b
+#LLM_MODEL=openai/qwen3-4b
 LLM_API_BASE=https://dashscope.aliyuncs.com/compatible-mode/v1
 LLM_API_KEY=
 LLM_MAX_TOKENS=16384

{xgae-0.2.0 → xgae-0.2.1}/CHANGELOG.md

@@ -1,4 +1,10 @@
 # Release Changelog
+## [0.2.1] - 2025-9-17
+### Added
+- Example Langgraph ReactAgent: add Chromadb for custom prompt RAG 
+### Modified
+- pyproject.toml: add [project.optional-dependencies] 'examples'
+
 ## [0.2.0] - 2025-9-10
 ### Added
 - Agent Engine release 0.2

{xgae-0.2.0 → xgae-0.2.1}/PKG-INFO

@@ -1,12 +1,16 @@
 Metadata-Version: 2.4
 Name: xgae
-Version: 0.2.0
+Version: 0.2.1
 Summary: Extreme General Agent Engine
 Requires-Python: >=3.13
 Requires-Dist: colorlog==6.9.0
 Requires-Dist: langchain-mcp-adapters==0.1.9
 Requires-Dist: langchain==0.3.27
 Requires-Dist: langfuse==2.60.9
-Requires-Dist: langgraph==0.6.5
 Requires-Dist: litellm==1.74.15
 Requires-Dist: mcp==1.13.0
+Provides-Extra: examples
+Requires-Dist: chromadb==1.1.0; extra == 'examples'
+Requires-Dist: fastembed==0.7.3; extra == 'examples'
+Requires-Dist: langchain-community==0.3.29; extra == 'examples'
+Requires-Dist: langgraph==0.6.5; extra == 'examples'

{xgae-0.2.0 → xgae-0.2.1}/pyproject.toml

@@ -1,19 +1,26 @@
 [project]
 name = "xgae"
-version = "0.2.0"
+version = "0.2.1"
 description = "Extreme General Agent Engine"
 readme = "README.md"
 requires-python = ">=3.13"
 dependencies = [
     "colorlog==6.9.0",
     "langchain-mcp-adapters==0.1.9",
-    "langgraph==0.6.5",
     "litellm==1.74.15",
     "mcp==1.13.0",
     "langfuse==2.60.9",
     "langchain==0.3.27",
 ]
 
+[project.optional-dependencies]
+examples = [
+    "langgraph==0.6.5",
+    "langchain-community==0.3.29",
+    "fastembed==0.7.3",
+    "chromadb==1.1.0",
+]
+
 [build-system]
 requires = ["hatchling"]
 build-backend = "hatchling.build"

xgae-0.2.1/src/examples/agent/langgraph/react/custom_prompt_rag.py

@@ -0,0 +1,68 @@
+import logging
+import os
+from typing import List, Optional
+
+os.environ['HF_ENDPOINT']="https://hf-mirror.com"
+os.environ['TOKENIZERS_PARALLELISM']="False"
+
+from langchain_community.embeddings.fastembed import FastEmbedEmbeddings
+from langchain_community.vectorstores import Chroma
+from langchain_core.documents import Document
+
+class CustomPromptRag:
+    PROMPT_SIMILAR_SCORE = 0.8
+
+    def __init__(self, emb_model_name: Optional[str] = "BAAI/bge-small-zh-v1.5"):
+        self.emb_model_name = emb_model_name
+        prompt_docs = []
+        # should read from DB, load all custom prompt or COT
+        prompt_docs.append(self._create_prompt_doc(
+            prompt_summary="Fault location and analysis of fault causes",
+            prompt_path="templates/example/fault_user_prompt.txt"
+        ))
+        self.vector_store = self._init_vector_store(prompt_docs)
+
+
+    def _create_prompt_doc(self, prompt_summary: str, prompt_path: str)-> Document:
+        return Document(
+            page_content=prompt_summary,
+            metadata={
+                "source": prompt_path,
+            }
+        )
+
+
+    def _init_vector_store(self, docs: List[Document]) -> Chroma:
+        # FastEmbedEmbeddings first time will download BAAI/bge-small-zh-v1.5 embedding model from HF
+        embeddings = FastEmbedEmbeddings(model_name=self.emb_model_name)
+        return Chroma.from_documents(documents=docs, embedding=embeddings, persist_directory=None)
+
+
+    def search_prompt(self, query:str)-> str | None:
+        prompt_path = None
+        results = self.vector_store.similarity_search_with_score(query=query, k=1)
+        if results and len(results) > 0:
+            doc, score = results[0]
+            if score > self.PROMPT_SIMILAR_SCORE:
+                logging.info(f"CustomPromptRag search: SIMILAR_SCORE: {score} > {self.PROMPT_SIMILAR_SCORE}, "
+                             f"\nquery: '{query}' \nprompt_summary: '{doc.page_content}'\n")
+            else:
+                prompt_path = doc.metadata['source']
+                logging.info(f"CustomPromptRag search: SIMILAR_SCORE: {score}, prompt_path: '{prompt_path}'")
+
+        return prompt_path
+
+if __name__ == "__main__":
+    from xgae.utils.setup_env import setup_logging
+
+    setup_logging()
+
+    custom_prompt_rag = CustomPromptRag()
+
+    querys = ["locate 10.2.3.4 fault and solution",
+             "5+7"]
+
+    for query in querys:
+        logging.info("*"*50)
+        logging.info(f"query: '{query}'")
+        custom_prompt_rag.search_prompt(query)

{xgae-0.2.0 → xgae-0.2.1}/src/examples/agent/langgraph/react/react_agent.py

@@ -22,12 +22,13 @@ from xgae.engine.task_engine import XGATaskEngine
 
 from examples.agent.langgraph.react.agent_base import AgentContext, TaskState, EvaluateResult
 from examples.agent.langgraph.react.result_eval_agent import TaskResultEvalAgent
+from examples.agent.langgraph.react.custom_prompt_rag import CustomPromptRag
 
 class XGAReactAgent:
     MAX_TASK_RETRY = 2
     QUALIFIED_RESULT_SCORE = 0.7
 
-    def __init__(self):
+    def __init__(self, use_prompt_rag: Optional[bool] = False):
         self.graph = None
 
         self.graph_config = None
@@ -37,6 +38,8 @@ class XGAReactAgent:
         self.tool_box = XGAMcpToolBox(custom_mcp_server_file="mcpservers/custom_servers.json")
         self.result_eval_agent = TaskResultEvalAgent()
 
+        if use_prompt_rag:
+            self.custom_prompt_rag = CustomPromptRag()
 
     async def _create_graph(self) -> StateGraph:
         try:
@@ -86,8 +89,10 @@ class XGAReactAgent:
 
 
     def _search_system_prompt(self, user_input: str) -> str:
-        # You should search RAG use user_input, fetch COT or Prompt for your business
-        system_prompt = None if "fault" not in user_input else read_file("templates/example/fault_user_prompt.txt")
+        if hasattr(self, 'custom_prompt_rag'):
+            system_prompt = self.custom_prompt_rag.search_prompt(user_input)
+        else:
+            system_prompt = None if "fault" not in user_input else read_file("templates/example/fault_user_prompt.txt")
         return system_prompt
 
 
@@ -145,7 +150,7 @@ class XGAReactAgent:
 
 
     def _select_custom_tools(self, system_prompt: str) -> list[str]:
-        # @todo select mcp tool based on system prompt in future
+        # @todo search mcp tool based on system prompt or user_input in future
         custom_tools = ["*"] if system_prompt  else []
         return custom_tools
 

{xgae-0.2.0 → xgae-0.2.1}/src/examples/agent/langgraph/react/result_eval_agent.py

@@ -27,6 +27,7 @@ class TaskResultEvalAgent:
 
         langfuse_metadata = self._create_llm_langfuse_meta(trace_id, session_id)
 
+        logging.info("TaskResultEvalAgent: LLM Eval result.....")
         response = await self.model_client.acompletion(messages, langfuse_metadata)
         response_text = await self.model_client.get_response_result(response)
 

{xgae-0.2.0 → xgae-0.2.1}/src/examples/agent/langgraph/react/run_react_agent.py

@@ -17,7 +17,7 @@ async def main():
     ]
 
     for user_input in user_inputs:
-        agent = XGAReactAgent()
+        agent = XGAReactAgent(use_prompt_rag=False)
         task_no += 1
         context: AgentContext = {
             'task_id': f"agent_task_{uuid4()}",   # can be set with request_id, must be unique

{xgae-0.2.0 → xgae-0.2.1}/src/examples/engine/run_custom_and_agent_tools.py

@@ -14,7 +14,7 @@ else:
     setup_logging()
 
 # Before Run Exec: uv run example-fault-tools --alarmtype=2  , uv run example-a2a-tools
-# If want to use real A2A agent tool, use xga-agent-tool project
+# If want to use real A2A agent tool, use xgaproxy project, uv run xga-a2a-proxy & uv run example-a2a-server
 
 async def main() -> None:
     tool_box = XGAMcpToolBox(custom_mcp_server_file="mcpservers/custom_servers.json")

{xgae-0.2.0 → xgae-0.2.1}/src/xgae/utils/llm_client.py

@@ -56,7 +56,7 @@ class LLMClient:
         env_llm_model           = os.getenv('LLM_MODEL', "openai/qwen3-235b-a22b")
         env_llm_api_key         = os.getenv('LLM_API_KEY')
         env_llm_api_base        = os.getenv('LLM_API_BASE', "https://dashscope.aliyuncs.com/compatible-mode/v1")
-        env_llm_max_tokens      = int(os.getenv('LLM_MAX_TOKENS', 16384))
+        env_llm_max_tokens      = int(os.getenv('LLM_MAX_TOKENS', 8192))
         env_llm_temperature     = float(os.getenv('LLM_TEMPERATURE', 0.7))
         env_llm_stream          = to_bool(os.getenv('LLM_STREAM', False))
         env_llm_enable_thinking = to_bool(os.getenv('LLM_ENABLE_THINKING', False))