LangChain: Address suggestions by CodeRabbit

Author: amotl

.github/workflows/ml-langchain.yml

@@ -2,7 +2,6 @@ name: LangChain
 
 on:
   pull_request:
-    branches: ~
     paths:
     - '.github/workflows/ml-langchain.yml'
     - 'topic/machine-learning/llm-langchain/**'

topic/machine-learning/llm-langchain/README.md

@@ -90,7 +90,7 @@ and [CrateDB].
   It is based on the previous notebook, and it illustrates how to use Vertex AI platform
   on Google Cloud for RAG pipeline.
 
-- `agent_with_mcp.py`
+- `agent_with_mcp.py` [![Open on GitHub](https://img.shields.io/badge/Open%20on-GitHub-lightgray?logo=GitHub)](agent_with_mcp.py)
 
   This example illustrates how to use LangGraph and the `langchain-mcp-adapters`
   package to implement an LLM agent that is connecting to the CrateDB MCP server.
@@ -173,7 +173,7 @@ pytest -k document_loader
 pytest -k "notebook and loader"
 ```
 
-To force a regeneration of the Jupyter Notebook, use the
+To force regeneration of Jupyter notebooks, use the
 `--nb-force-regen` option.
 ```shell
 pytest -k document_loader --nb-force-regen

topic/machine-learning/llm-langchain/agent_with_mcp.py

@@ -24,6 +24,7 @@
 python agent_with_mcp.py
 """
 import asyncio
+import os
 
 from cratedb_about.instruction import GeneralInstructions
 from langchain_mcp_adapters.client import MultiServerMCPClient
@@ -46,7 +47,7 @@ async def amain():
         prompt=GeneralInstructions().render(),
     )
 
-    QUERY_STR = "What is the average value for sensor 1?"
+    QUERY_STR = os.getenv("DEMO_QUERY", "What is the average value for sensor 1?")
     response = await agent.ainvoke({"messages": QUERY_STR})
     answer = response["messages"][-1].content