Functionality

This module is designed to identify stocks based on the user's question and provide a response by loading each fact sheet into the context window of the language model and executing the requested task. The input is a detailed question from the user related to information within the document.
To aid in identifying the correct document and to correct for spelling mistakes, information from the stock universe .csv file will be used.

Example inputs:

• "Please provide me the investment rationales for the above-mentioned documents."

• "Please provide me with a compelling story about NVIDIA for my client."

The user has two options:

1. Refer to stocks extracted by a table in the previous chat interaction or mentioned in a previous conversation.

2. Directly mention the relevant stocks in the user question.

The output is the rationale from the document for each stock, provided as answers to the user's questions.

Input

1. A user question related to detailed information within the document.

2. Uploaded .csv table and connected documents.
   Examples:

   1. CSV Table Demo_stocks.csv

      Open

      

   2. Document CH0012221716-factsheet.pdf since config for identification is set to "stockFileName": "$ISIN$-factsheet.pdf".

Output

The rationale from the document for each stock, provided as answers to the user's questions.

Open

Configuration settings (technical)

This section contains all configurable parameters as seen in the JSON.

Here is the updated input formatted into a table with the specified types and similar text for prompts:

Configuration settings (technical)

This section contains all configurable parameters as seen in the JSON.

General parameters

Prompts