# Todo

* [ ] Add `Evaluate` buttons to `System Message` and `Initial System Message` for convenient message evaluation.
* [ ] Implement `Map Reduce` using the guide provided at https://langchain-ai.github.io/langgraph/how-tos/map-reduce/ for meta-prompting with batch data.
* [ ] Fine-tune a model/LoRa for Meta Prompt.
  * [ ] Collect/create a data set for the fine-tuning.
  * [ ] Create a benchmark for Meta Prompt.
* [ ] `Clear` button for Streamlit UI.
* [ ] `Validate` button for `Prompt` tab.
* [ ] `Diff` button for `Prompt` tab to show the difference between the expected output and the actual output.