diff --git a/README.md b/README.md
index 94c8ee9..a4e3a4e 100644
--- a/README.md
+++ b/README.md
@@ -87,7 +87,7 @@ From there, it's easy to implement popular design patterns like ([Multi-](https:
 | [Code Generator](https://github.com/The-Pocket/PocketFlow/tree/main/cookbook/pocketflow-code-generator) | ★☆☆ <sup>*Beginner*</sup> | Generate test cases, implement solutions, and iteratively improve code |
 | [MCP](https://github.com/The-Pocket/PocketFlow/tree/main/cookbook/pocketflow-mcp) |  ★☆☆ <sup>*Beginner*</sup> |  Agent using Model Context Protocol for numerical operations |
 | [A2A](https://github.com/The-Pocket/PocketFlow/tree/main/cookbook/pocketflow-a2a) |  ★☆☆ <sup>*Beginner*</sup> | Agent wrapped with A2A protocol for inter-agent communication |
-| [Streamlit HITL](https://github.com/The-Pocket/PocketFlow/tree/main/cookbook/pocketflow-streamlit-hitl) | ★☆☆ <sup>*Beginner*</sup> | Streamlit app for human-in-the-loop review |
+| [Streamlit HITL](https://github.com/The-Pocket/PocketFlow/tree/main/cookbook/pocketflow-streamlit-hitl) | ★☆☆ <sup>*Beginner*</sup> | Streamlit app for human-in-the-loop image generation |
 | [FastAPI HITL](https://github.com/The-Pocket/PocketFlow/tree/main/cookbook/pocketflow-fastapi-hitl)    | ★☆☆ <sup>*Beginner*</sup> | FastAPI app  for async human review loop with SSE |
 | [Voice Chat](https://github.com/The-Pocket/PocketFlow/tree/main/cookbook/pocketflow-voice-chat) | ★☆☆ <sup>*Beginner*</sup> | An interactive voice chat application with VAD, STT, LLM, and TTS. |
 
diff --git a/cookbook/pocketflow-streamlit-hitl/README.md b/cookbook/pocketflow-streamlit-hitl/README.md
index 52d76de..3af3273 100644
--- a/cookbook/pocketflow-streamlit-hitl/README.md
+++ b/cookbook/pocketflow-streamlit-hitl/README.md
@@ -2,6 +2,12 @@
 
 Human-in-the-Loop (HITL) image generation application using PocketFlow and Streamlit. Enter text prompts, generate images with OpenAI, and approve/regenerate results.
 
+<p align="center">
+  <img 
+    src="./assets/banner.png" width="800"
+  />
+</p>
+
 ## Features
 
 -   **Image Generation:** Uses OpenAI's `gpt-image-1` model to generate images from text prompts
diff --git a/cookbook/pocketflow-streamlit-hitl/assets/banner.png b/cookbook/pocketflow-streamlit-hitl/assets/banner.png
new file mode 100644
index 0000000..8217686
Binary files /dev/null and b/cookbook/pocketflow-streamlit-hitl/assets/banner.png differ
diff --git a/cookbook/pocketflow-streamlit-hitl/utils/process_task.py b/cookbook/pocketflow-streamlit-hitl/utils/process_task.py
deleted file mode 100644
index 1cc13b0..0000000
--- a/cookbook/pocketflow-streamlit-hitl/utils/process_task.py
+++ /dev/null
@@ -1,24 +0,0 @@
-import time
-
-def process_task(task_input: str) -> str:
-    """
-    Simulates processing the input, potentially calling an LLM.
-    Replace this with your actual task logic.
-    """
-    print(f"Processing task: {task_input[:50]}...")
-
-    result = f"Dummy rephrased text for the following input: {task_input}"
-
-    # Simulate some work
-    time.sleep(2) 
-    return result
-
-if __name__ == "__main__":
-    test_input = "This is a test input for the processing task."
-    print(f"Input: {test_input}")
-    output = process_task(test_input)
-    print(f"Output: {output}")
-
-# We don't need a separate utils/call_llm.py for this minimal example,
-# but you would add it here if ProcessNode used an LLM.
-