-
Notifications
You must be signed in to change notification settings - Fork 95
Chat
This wiki page provides a comprehensive overview of the Chat feature in Knowledge, explaining how it enhances user interaction with Sources and Projects. The Chat feature is available in two forms: the Chat view and individual Source chats accessible through the Source details page.
The Chat feature allows users to communicate with agents representing their Sources and Projects. These agents facilitate learning by providing valuable insights and recommendations. The Chat feature is accessible in two ways:
- Chat View: Offers a centralized location for all individual chats with Sources and the active Project.
- Source Chat: Accessible through the Source details page, allowing users to chat with a specific Source.
The Chat view displays all conversations between the user, the active Project agent, and the Source agent for all inherited Sources. This view enables users to ask questions and gather information about all Sources in the active Project. The Chat view provides the following benefits:
- Centralized Communication: Users can easily access all conversations in one place, making it more convenient to manage and review information.
- Improved Learning: Project agents review conversations between Source agents and users, guiding the learning process and suggesting next steps based on user interactions.
- Enhanced Interaction: Users can ask questions and receive tailored information, including summaries, key topics, and related Source recommendations.
In addition to the Chat view, each Source has its own chat accessible through the Source details page. Users can engage with individual Source agents to gather specific information about the Source. The Source chat provides the following benefits:
- Focused Learning: Users can delve deeper into a particular Source, receiving targeted information and insights from the Source agent.
- Personalized Interaction: Source agents can provide customized summaries and key topic extraction based on the user's questions and interests.
- Quick Access: The Source chat is conveniently located within the Source details page, allowing users to chat with the Source agent while reviewing the Source's metadata.
Due to technical limitations, Knowledge is currently only able to read (and therefore chat about) the following file types (PDF, Code, and plain-text):
.pdf
.txt
.c
.h
.cpp
.hpp
.cs
.go
.rs
.sh
.py
.js
.ts
.java
.html
.htm
.css
.scss
Dockerfile
.json
.xml
.yaml
.yml
.md
.jsx
.rb
.rake
.pl
.php
.sql
.swift
.kt
.jl
.graphql
.erl
.d
.curl
.diff
.liquid
.r
To utilize the Chat view or Source chat in Knowledge, users need to sign up for an OpenAI API key and enter it into the Knowledge application. This section provides a step-by-step guide on how to obtain an OpenAI API key.
- Visit the OpenAI website and click on "Sign Up" or "Sign In" if you already have an account.
- Complete the registration process and log in to your account.
- Navigate to the "View API Keys" section on your OpenAI Dashboard.
- Click on "Create a new secret key" and follow the prompts to generate your unique API key.
- Copy the generated API key to your clipboard, as you will need it to enable the Chat features in Knowledge.
Once you have your API key, perform the following steps:
- Launch the Knowledge application.
- Click on the Chat icon on the left sidebar.
- A dialog will appear asking for your API Key. Paste the API key and click "Submit".
- If the API key works, the Chat view will be displayed, and the Source chat will be enabled whenever you view the details page of any individual Source.
Note that your API key will be encrypted and written to a file on your computer.
The Chat feature in Knowledge enhances user interaction with Sources and Projects, facilitating learning and information gathering. With the Chat view and individual Source chats, users can conveniently communicate with agents representing their Sources and Projects, receiving valuable insights and recommendations tailored to their needs. By leveraging the Chat feature, users can optimize their learning experience within the Knowledge application.