Needs light modification to run with Ollama

[gbetous]

To make the (amazing !) tutorial work with Ollama (it works great with qwen3-coder) I have to modify every agent to remove those two lines.

diff --git a/agents/s01_agent_loop.py b/agents/s01_agent_loop.py
index 3e6c913..893a3d7 100644
--- a/agents/s01_agent_loop.py
+++ b/agents/s01_agent_loop.py
@@ -31,9 +31,6 @@ from dotenv import load_dotenv
 
 load_dotenv(override=True)
 
-if os.getenv("ANTHROPIC_BASE_URL"):
-    os.environ.pop("ANTHROPIC_AUTH_TOKEN", None)
-
 client = Anthropic(base_url=os.getenv("ANTHROPIC_BASE_URL"))
 MODEL = os.environ["MODEL_ID"]

What is their purpose ? I can prepare a PR to allow usage on both Claude and Ollama but I don't understand the basic need.

Thanks !

FYI, my .env content:

ANTHROPIC_AUTH_TOKEN=ollama
ANTHROPIC_API_KEY=""
ANTHROPIC_BASE_URL=http://localhost:11434
MODEL_ID=qwen3-coder