[ ] Ensure that multi-agent and single-agent are using the same query engine tool for TS and Python
[ ] Enhance the query engine tool to have two tools if LlamaCloud is used (one using chunks, one using whole docs) (see notebook). The code should be copied if possible (no run-time check for LlamaCloud)
[ ] Add use case to simple mode (Replace Multi-Agent Report Gen) - as in the notebook
based on Jupyter notebook: https://github.com/run-llama/llamacloud-demo/blob/main/examples/report_generation/report_generation.ipynb
We can use the pre-configured LlamaCloudIndex from https://github.com/run-llama/llamacloud-demo/blob/main/examples/form_filling/Form_Filling_10K_SEC.ipynb for this task as it contains the 10k docs (this is validated)
Tasks: