Skip to content
Merged
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
38 changes: 22 additions & 16 deletions README.md
Original file line number Diff line number Diff line change
@@ -1,29 +1,35 @@
# chergpt-basic: custom chat assistant
No login, vanilla ChatGPT-like interface with scaffolded code for immediate deployment
# Chergpt-Basic: Custom Chat Assistant

Chergpt-Basic is a simplified, login-free ChatGPT-like interface designed for quick deployment. It allows users to interact with a chat assistant based on OpenAI's GPT-3.5 model. The README provides clear instructions on setting up the backend, customization, and potential use cases.

## Deployment
- https://chergpt-physics-lookang.streamlit.app/ from my own forked GitHub
- https://chergpt-physics-cpdd.streamlit.app/ from Kah How's deployment

## Features
- [x] Set custom instructions to guide student interactions <br>
- [WIP] Trigger immediate summaries of learning insights at the end of the session <br>
- [x] Set custom instructions to guide student interactions.
- [WIP] Trigger immediate summaries of learning insights at the end of the session.

Need to gate access with a simple global password? See [here](https://docs.streamlit.io/knowledge-base/deploy/authentication-without-sso)
Need to secure access with a simple global password? Check the [authentication guide](https://docs.streamlit.io/knowledge-base/deploy/authentication-without-sso).

## Use cases
The most straightforward use of custom instructions is to create user personas for students to interact with. <br>
E.g. historical figures, patients with particular medical conditions
## Use Cases
Chergpt-Basic enables a variety of use cases, with the primary one being the creation of user personas for students to interact with. Examples include historical figures or patients with specific medical conditions.

## How to setup backend
We recommend using **NeonDB**
You can follow instructions on NeonDB [here](https://start.open.gov.sg/docs/getting-started/prerequisites) <br>

Your cluster will be created in a few seconds and the Create SQL user dialog will display.

For either (A) or (B), copy the URL and put it as `DB_CONNECTION = "{DB connection String}"` under `secrets.toml`
## Secrets.toml File Structure
In the `secrets.toml` file, specify three items: you have to change it to your own

## secrets.toml file structure
You will need to specify 3 items:<br>
```
OPENAI_API_KEY = "{api_key}" #See [here](https://teachertech.beehiiv.com/p/api-openai) if you unsure on how to create<br>
DB_CONNECTION = "{db_connection string}" #Refer to instructions above on **How to setup backend** if unsure
ADMIN_PASSWORD = "{any password you want}" #This is used to gate access on how can update custom instructions
```toml
OPENAI_API_KEY = "sk-Co6R5h4RcT....................................."
ADMIN_PASSWORD= "passwordyourchoice"
DB_CONNECTION = "postgresql://chergpt:PGPASSWORD@ep-weathered-shadow-a1uz6hly.ap-southeast-1.aws.neon.tech/neondb?sslmode=require"
```
- See [here](https://teachertech.beehiiv.com/p/api-openai) if you're unsure how to create it.
- Used to gate access for updating custom instructions.
- Refer to instructions above on **How to Set Up Backend** if unsure.

Feel free to explore and enhance the functionality of Chergpt-Basic according to your specific needs!