Merge pull request #51 from ks6088ts-labs/feature/issue-46_streamlit

add Streamlit PoC demo

Author: ks6088ts

Makefile

@@ -127,3 +127,8 @@ fastapi: ## run FastAPI in production mode
 		--port 8000 \
 		--workers 4 \
 		template_langgraph/services/fastapis/main.py
+
+.PHONY: streamlit
+streamlit: ## run Streamlit
+	uv run streamlit run \
+		template_langgraph/services/streamlits/main.py

docs/faq.ja.md

@@ -211,6 +211,19 @@ uv run fastapi run \
 
 ![fastapi.png](./images/fastapi.png)
 
+### オプション 5: Streamlit（PoC デモ）
+
+Streamlit アプリを実行して、エージェントを Web インターフェースとして公開：
+
+```shell
+uv run streamlit run \
+  template_langgraph/services/streamlits/main.py
+```
+
+Streamlit アプリのデモ：
+
+[![streamlit.png](./images/streamlit.png)](https://youtu.be/z7QhEsqIGAk)
+
 ## 実演されている主要概念
 
 ### 1. **ReAct パターン**（推論 + 行動）

docs/images/streamlit.png

@@ -211,6 +211,19 @@ This allows you to interact with the agent programmatically via HTTP requests.
 
 ![fastapi.png](./images/fastapi.png)
 
+### Option 5: Streamlit (PoC demo)
+
+Run the Streamlit app to expose the agent as a web interface:
+
+```shell
+uv run streamlit run \
+  template_langgraph/services/streamlits/main.py
+```
+
+Demonstration of the Streamlit app:
+
+[![streamlit.png](./images/streamlit.png)](https://youtu.be/z7QhEsqIGAk)
+
 ## Key Concepts Demonstrated
 
 ### 1. **ReAct Pattern** (Reasoning + Acting)

docs/index.ja.md

@@ -6,6 +6,7 @@
 - [LangGraph の(LLM なし)Human-in-the-loop を試してみた](https://qiita.com/te_yama/items/db38201af60dec76384d)
 - [🤖 LangGraph Multi-Agent Supervisor](https://github.com/langchain-ai/langgraph-supervisor-py)
 - [Software Design 誌「実践 LLM アプリケーション開発」第 24 回サンプルコード](https://github.com/mahm/softwaredesign-llm-application/tree/main/24)
+- [Streamlit](https://python.langchain.com/docs/integrations/callbacks/streamlit/)
 
 ### Sample Codes
 

docs/index.md

@@ -21,6 +21,7 @@ dependencies = [
     "pypdf>=5.9.0",
     "python-dotenv>=1.1.0",
     "qdrant-client>=1.15.1",
+    "streamlit>=1.48.0",
     "typer>=0.16.0",
 ]
 

docs/references.ja.md

@@ -0,0 +1,4 @@
+import streamlit as st
+
+st.title("template-streamlit")
+st.info("Select a code sample from the sidebar to run it")