https://github.com/hexastack/hexabot-helper-chatgpt
The OpenAI ChatGPT Helper Extension for Hexabot Chatbot / Agent Builder to enable the LLM Capability
https://github.com/hexastack/hexabot-helper-chatgpt
agent agentic-ai ai chatbot chatbot-framework chatbots chatgpt chatgpt-api hexabot large-language-model large-language-models llm openai openai-api openai-api-chatbot openai-chatgpt rag
Last synced: about 1 month ago
JSON representation
The OpenAI ChatGPT Helper Extension for Hexabot Chatbot / Agent Builder to enable the LLM Capability
- Host: GitHub
- URL: https://github.com/hexastack/hexabot-helper-chatgpt
- Owner: Hexastack
- Created: 2024-10-23T09:33:43.000Z (8 months ago)
- Default Branch: main
- Last Pushed: 2025-02-14T09:33:28.000Z (4 months ago)
- Last Synced: 2025-03-31T10:42:30.295Z (3 months ago)
- Topics: agent, agentic-ai, ai, chatbot, chatbot-framework, chatbots, chatgpt, chatgpt-api, hexabot, large-language-model, large-language-models, llm, openai, openai-api, openai-api-chatbot, openai-chatgpt, rag
- Language: TypeScript
- Homepage: https://docs.hexabot.ai
- Size: 37.1 KB
- Stars: 2
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Hexabot ChatGPT Helper Extension
The **Hexabot ChatGPT Helper Extension** is a utility class designed to facilitate interaction with the ChatGPT API from other Hexabot extensions (such as plugins, channels, etc.). This helper allows developers to easily invoke the ChatGPT API and integrate natural language understanding and response generation into Hexabot's chatbot builder.
[Hexabot](https://hexabot.ai/) is an open-source chatbot / agent solution that allows users to create and manage AI-powered, multi-channel, and multilingual chatbots with ease. If you would like to learn more, please visit the [official github repo](https://github.com/Hexastack/Hexabot/).
## Features
- **API Integration**: Seamlessly connect to OpenAI's ChatGPT API, enabling other extensions within Hexabot to access ChatGPT's capabilities.
- **Configurable Settings**: Configure parameters like model type, temperature, maximum tokens, penalties, and more for customized behavior.
- **Easy Integration**: Use as a helper utility to invoke the ChatGPT API from any other extension within Hexabot.
- **Flexible Options**: Supports various options such as response format, stop sequences, log probabilities, and more to customize the behavior of ChatGPT.
## Installation
To use the ChatGPT Helper Extension within Hexabot, follow these steps:
```
cd ~/projects/my-chatbot
npm install hexabot-helper-chatgpt
hexabot dev
```
Make sure you have the appropriate access credentials for the ChatGPT API.
## Usage
The ChatGPT Helper can be used to generate responses based on user input or integrate into more complex workflows involving conversation history and system prompts. Here's an overview of how to use this helper:
### Settings
The extension provides configurable settings that can be adjusted to suit your needs. Below are the available settings:
- **API Token**: The API key to authenticate with the ChatGPT API.
- **Model**: Specifies which model to use (default: `gpt-4o-mini`).
- **Temperature**: Controls the randomness of the output (default: `0.8`).
- **Max Completion Tokens**: The maximum number of tokens for the completion (default: `1000`).
- **Frequency Penalty**: Controls the repetition of words (default: `0`).
- **Presence Penalty**: Controls the introduction of new topics (default: `0`).
- **Tool Choice**: Option for using specific tools (`none`, `auto`, or `required`).
- **Response Format**: Format of the response (`text` or `json`).
These settings can be customized using the Hexabot admin interface or programmatically via the Hexabot API.
### 1. Generate a Response
You can use the helper to generate responses from the model based on user input:
```typescript
const response = await chatGptHelper.generateResponse(prompt, model, systemPrompt, options);
console.log(response);
```
##### Example
```typescript
const response = await chatGptHelper.generateResponse(
'Where is Paris located?',
'gpt-4o-mini',
'You are a tourist assistant',
{...}
);
```
#### Parameters:
- **prompt:** The user input or query for which a response is generated.
- **model:** Specifies the language model to be used (e.g., "'gpt-4o-mini").
- **systemPrompt:** A system instruction that sets the behavior or tone of the model's response.
- **options:** Additional settings, such as temperature, max completion tokens, or other settings.
### 2. Generate a Chat Completion
This method uses conversation history to generate a contextual response:
```typescript
const response = await chatGptHelper.generateChatCompletion(prompt, model, systemPrompt, history, options);
console.log(response);
```
#### Parameters:
- **prompt:** The latest user input in a conversation.
- **model:** The name of the language model used for generating responses.
- **systemPrompt:** Instructions that help guide the model's response.
- **history:** A conversation history consisting of previous exchanges, ensuring context-aware responses.
- **options:** Custom parameters for model behavior, such as response length, temperature, logit bias or stop sequences.
### 3. Generate a Structured Response
```typescript
const response = await chatGptHelper.generateStructuredResponse(prompt, model, systemPrompt, schema, options);
console.log(response);
```
##### Example
```typescript
const response = await chatGptHelper.generateResponse(
'Where is Paris located?',
'gpt-4o-mini',
'You are a tourist assistant',
{
type: 'string',
description: 'Country of the input destination'
},
{...}
);
```
#### Parameters
- **prompt:** The query or instruction provided to the model.
- **model:** The model used for structured response generation.
- **systemPrompt:** Guidelines for formatting or structuring the response.
- **schema:** Defines the expected structure of the response.
- **options:** Additional configurations like response constraints or output format.
### Example Integration
To use the ChatGPT Helper, simply inject the `ChatGptLlmHelper` class and use it as shown below:
```typescript
const chatGptHelper = this.helperService.use(
HelperType.LLM,
ChatGptLlmHelper,
);
// ...
const text = await chatGptHelper.generateChatCompletion(
context.text,
args.model,
systemPrompt,
history,
{
...options,
logit_bias: JSON.parse(options.logit_bias) as Record || {},
user: context.user.id,
},
);
```
## Contributing
We welcome contributions from the community! Whether you want to report a bug, suggest new features, or submit a pull request, your input is valuable to us.
Please refer to our contribution policy first : [How to contribute to Hexabot](https://github.com/Hexastack/Hexabot/blob/main/CONTRIBUTING.md)
[](https://github.com/Hexastack/Hexabot/blob/main/CODE_OF_CONDUCT.md)
Feel free to join us on [Discord](https://discord.gg/rNb9t2MFkG)
## License
This software is licensed under the GNU Affero General Public License v3.0 (AGPLv3) with the following additional terms:
1. The name "Hexabot" is a trademark of Hexastack. You may not use this name in derivative works without express written permission.
2. All derivative works must include clear attribution to the original creator and software, Hexastack and Hexabot, in a prominent location (e.g., in the software's "About" section, documentation, and README file).
---
_Happy Chatbot Building!_