Merge pull request #106 from ks6088ts-labs/feature/issue-105_parallel-rag-agent

update from parallel processor agent to parallel rag agent

Author: ks6088ts

docs/index.ja.md

@@ -89,7 +89,7 @@ uv run python scripts/elasticsearch_operator.py create-index \
 # Azure AI Search
 make create-ai-search-index
 
-# Azure Cosmos DB  
+# Azure Cosmos DB
 make create-cosmosdb-index
 
 # またはオペレータースクリプトを直接使用:
@@ -129,7 +129,7 @@ make create-cosmosdb-index
 
 - `weather_agent.py` — シンプルなツール呼び出しエージェント。モック天気検索ツールを使った基本的な ReAct パターン。主要概念: ツール呼び出し、基本エージェントパターン。
 - `multi_agent.py` — マルチエージェント協調。転送機能を使ったエージェント間の引き渡しを実演。主要概念: エージェント協調、ワークフロー転送。
-- `parallel_processor_agent/` — 並列実行タスク分解。目標をタスクに分解し並列処理。主要概念: 並列処理、タスク分解、Send 操作。
+- `parallel_rag_agent/` — 並列実行タスク分解。目標をタスクに分解し並列処理。主要概念: 並列処理、タスク分解、Send 操作。
 
 ### サポートモジュール
 
@@ -287,11 +287,11 @@ uv run python scripts/demo_agents_operator.py multi-agent \
   --verbose
 ```
 
-- Parallel processor agent（タスク分解）:
+- Parallel rag agent（タスク分解）:
 
 ```shell
-uv run python scripts/demo_agents_operator.py parallel-processor-agent \
-  --goal "ソフトウェア会社立ち上げのための情報収集戦略を計画する" \
+uv run python scripts/demo_agents_operator.py parallel-rag-agent \
+  --query "ソフトウェア会社立ち上げのための情報収集戦略を計画する" \
   --verbose
 ```
 
@@ -377,7 +377,7 @@ make mcp-inspector
 ## 次のステップ
 
 1. **基本から始める**: `kabuto_helpdesk_agent`の例を実行
-2. **デモエージェントを試す**: `weather_agent`、`multi_agent`、`parallel_processor_agent`でシンプルなパターンを探索
+2. **デモエージェントを試す**: `weather_agent`、`multi_agent`、`parallel_rag_agent`でシンプルなパターンを探索
 3. **実装を理解する**: `chat_with_tools_agent`と比較
 4. **高度なパターンを探索**: タスク分解器とスーパーバイザーエージェントを試す
 5. **独自のものを構築**: このテンプレートをあなたのユースケースの出発点として使用
@@ -404,6 +404,6 @@ uv run python scripts/otel_operator.py run -q "health check" -v
 3. **構造化出力エージェント** - フォーマットされたデータの返却（`issue_formatter_agent`）
 4. **計画エージェント** - 複雑なタスクの分解（`task_decomposer_agent`）
 5. **マルチエージェントシステム** - 複数エージェントの協調（`supervisor_agent`、`multi_agent`）
-6. **並列処理** - 同時タスク実行（`parallel_processor_agent`）
+6. **並列処理** - 同時タスク実行（`parallel_rag_agent`）
 
 各パターンは、いつどのように使用するかを理解するのに役立つ明確な例と文書で実装されています。

docs/index.md

@@ -89,7 +89,7 @@ uv run python scripts/elasticsearch_operator.py create-index \
 # Azure AI Search
 make create-ai-search-index
 
-# Azure Cosmos DB  
+# Azure Cosmos DB
 make create-cosmosdb-index
 
 # Or use the operator scripts directly:
@@ -129,7 +129,7 @@ Additional simple agents for learning and demonstration:
 
 - `weather_agent.py` — Simple tool-calling agent. Basic ReAct pattern with mock weather search tool. Key: tool calling, basic agent pattern.
 - `multi_agent.py` — Multi-agent coordination. Demonstrates agent-to-agent handoff using transfer functions. Key: agent coordination, workflow transfer.
-- `parallel_processor_agent/` — Task decomposition with parallel execution. Breaks down goals into tasks and processes them in parallel. Key: parallel processing, task decomposition, Send operations.
+- `parallel_rag_agent/` — Task decomposition with parallel execution. Breaks down querys into tasks and processes them in parallel. Key: parallel processing, task decomposition, Send operations.
 
 ### Supporting Modules
 
@@ -287,11 +287,11 @@ uv run python scripts/demo_agents_operator.py multi-agent \
   --verbose
 ```
 
-- Parallel processor agent (task decomposition):
+- Parallel rag agent (task decomposition):
 
 ```shell
-uv run python scripts/demo_agents_operator.py parallel-processor-agent \
-  --goal "Plan information gathering strategy for launching a software company" \
+uv run python scripts/demo_agents_operator.py parallel-rag-agent \
+  --query "Plan information gathering strategy for launching a software company" \
   --verbose
 ```
 
@@ -379,7 +379,7 @@ make mcp-inspector
 ## Next Steps
 
 1. **Start with the basics**: Run the `kabuto_helpdesk_agent` example
-2. **Try demo agents**: Explore the simple patterns with `weather_agent`, `multi_agent`, and `parallel_processor_agent`
+2. **Try demo agents**: Explore the simple patterns with `weather_agent`, `multi_agent`, and `parallel_rag_agent`
 3. **Understand the implementation**: Compare it with `chat_with_tools_agent`
 4. **Explore advanced patterns**: Try the task decomposer and supervisor agents
 5. **Build your own**: Use this template as a starting point for your use case
@@ -406,6 +406,6 @@ This template demonstrates several proven agent architectures:
 3. **Structured Output Agent** - Returning formatted data (`issue_formatter_agent`)
 4. **Planning Agent** - Breaking down complex tasks (`task_decomposer_agent`)
 5. **Multi-Agent Systems** - Coordinating multiple agents (`supervisor_agent`, `multi_agent`)
-6. **Parallel Processing** - Concurrent task execution (`parallel_processor_agent`)
+6. **Parallel Processing** - Concurrent task execution (`parallel_rag_agent`)
 
 Each pattern is implemented with clear examples and documentation to help you understand when and how to use them.

langgraph.json

@@ -1,10 +1,8 @@
 {
-  "dependencies": [
-    "."
-  ],
+  "dependencies": ["."],
   "graphs": {
     "chat_with_tools_agent": "template_langgraph.agents.chat_with_tools_agent.agent:graph",
-    "demo_agents_parallel_processor_agent": "template_langgraph.agents.demo_agents.parallel_processor_agent.agent:graph",
+    "demo_agents_parallel_rag_agent": "template_langgraph.agents.demo_agents.parallel_rag_agent.agent:graph",
     "demo_agents_multi_agent": "template_langgraph.agents.demo_agents.multi_agent:graph",
     "demo_agents_weather_agent": "template_langgraph.agents.demo_agents.weather_agent:graph",
     "image_classifier_agent": "template_langgraph.agents.image_classifier_agent.agent:graph",
@@ -15,4 +13,4 @@
     "task_decomposer_agent": "template_langgraph.agents.task_decomposer_agent.agent:graph"
   },
   "env": ".env"
-}
+}

scripts/demo_agents_operator.py

@@ -4,7 +4,7 @@
 from dotenv import load_dotenv
 
 from template_langgraph.agents.demo_agents.multi_agent import graph as multi_agent_graph
-from template_langgraph.agents.demo_agents.parallel_processor_agent.agent import graph as parallel_processor_agent_graph
+from template_langgraph.agents.demo_agents.parallel_rag_agent.agent import graph as parallel_rag_agent_graph
 from template_langgraph.agents.demo_agents.weather_agent import graph as weather_agent_graph
 from template_langgraph.loggers import get_logger
 
@@ -74,12 +74,12 @@ def multi_agent(
 
 
 @app.command()
-def parallel_processor_agent(
-    goal: str = typer.Option(
-        "ソフトウェアシステム開発会社を立ち上げる戦略を立てるための情報収集をしたい",
-        "--goal",
-        "-g",
-        help="The goal to decompose into tasks",
+def parallel_rag_agent(
+    query: str = typer.Option(
+        "KABUTO のシステム概要やトラブルシュート事例を多種多様な情報ソースから回答して",
+        "--query",
+        "-q",
+        help="The query to decompose into tasks",
     ),
     verbose: bool = typer.Option(
         False,
@@ -91,9 +91,9 @@ def parallel_processor_agent(
     if verbose:
         logger.setLevel(logging.DEBUG)
 
-    for event in parallel_processor_agent_graph.stream(
+    for event in parallel_rag_agent_graph.stream(
         input={
-            "goal": goal,
+            "query": query,
         },
         debug=True,
     ):

template_langgraph/agents/demo_agents/parallel_processor_agent/agent.py

@@ -0,0 +1,54 @@
+from langchain_core.language_models.chat_models import BaseChatModel
+from langchain_core.tools.base import BaseTool
+from langgraph.graph import StateGraph
+from langgraph.graph.state import CompiledStateGraph
+
+from template_langgraph.agents.demo_agents.parallel_rag_agent.models import (
+    ParallelRagAgentInputState,
+    ParallelRagAgentOutputState,
+    ParallelRagAgentState,
+)
+from template_langgraph.agents.demo_agents.parallel_rag_agent.nodes.decompose_tasks import DecomposeTasks
+from template_langgraph.agents.demo_agents.parallel_rag_agent.nodes.run_task import RunTask
+from template_langgraph.agents.demo_agents.parallel_rag_agent.nodes.summarize_results import SummarizeResults
+from template_langgraph.llms.azure_openais import AzureOpenAiWrapper
+from template_langgraph.tools.common import get_default_tools
+
+
+class ParallelRagAgent:
+    def __init__(
+        self,
+        llm: BaseChatModel,
+        tools: list[BaseTool],
+    ):
+        self.llm = llm
+        self.decompose_tasks = DecomposeTasks(
+            llm=llm,
+            tools=tools,
+        )
+        self.run_task = RunTask(
+            llm=llm,
+            tools=tools,
+        )
+        self.summarize_results = SummarizeResults()
+
+    def create_graph(self) -> CompiledStateGraph:
+        workflow = StateGraph(
+            state_schema=ParallelRagAgentState,
+            input_schema=ParallelRagAgentInputState,
+            output_schema=ParallelRagAgentOutputState,
+        )
+        workflow.add_node("decompose_tasks", self.decompose_tasks)
+        workflow.add_node("run_task", self.run_task)
+        workflow.add_node("summarize_results", self.summarize_results)
+
+        workflow.add_edge("run_task", "summarize_results")
+        workflow.set_entry_point("decompose_tasks")
+        workflow.set_finish_point("summarize_results")
+        return workflow.compile()
+
+
+graph = ParallelRagAgent(
+    llm=AzureOpenAiWrapper().chat_model,
+    tools=get_default_tools(),
+).create_graph()