Step-thru Debugging your LangGraphs

Most of the details required are already covered in langgraph tutorial here: https://langchain-ai.github.io/langgraph/tutorials/langgraph-platform/local-server/

Key points covered there are:

1. Install LangGraph CLI

   pip install --upgrade "langgraph-cli[inmem]"
   

2. Install the dependencies

   pip install -e .
   

3. Use the CLI to run langGraph server

   langgraph dev
   

4. You may now use LangGraph Studio Web UI at: https://smith.langchain.com/studio/?baseUrl=http://locahost:2024

Here are the additions to enable step-thru-debugging:

1. Install debugpy in your virtual environment for the langgraph project

   pip install debugpy
   

2. Start langgraph API server in debug mode. In the following we are telling langgraph-cli to allow clients to listen on Port 2025 for debugger events

   langgraph dev --debug-port 2025
   

3. Attach to debugger from your IDE. E.g., in here’s my VS Code launch config:

   {
       "version": "0.2.0",
       "configurations": [
   
           {
               "name": "Python Debugger: Remote Attach",
               "type": "debugpy",
               "request": "attach",
               "connect": {
                   "host": "localhost",
                   "port": 2025
               },
               "pathMappings": [
                   {
                       "localRoot": "${workspaceFolder}",
                       "remoteRoot": "."
                   }
               ],
               "justMyCode": false
           }
       ]
   }
   

4. I can now put a breakpoint on any of my langgraph code files and step-thru the execution!