An API assistant with ChatGPT

APIGPT is an API assistant that uses natural language understanding to help users interact with APIs through conversation, including generating example code and checking parameter errors.

The design purpose of APIGPT is to simplify the process of using APIs. Users can describe their needs in natural language, and APIGPT will understand their needs based on the API documentation and provide the appropriate response. APIGPT learns about APIs by reading online API documentation or Github document repositories, and ultimately serves users through conversation.

APIGPT is based on the natural language understanding technology called ChatGPT. ChatGPT is a large-scale pre-trained language model that can convert natural language input into a format that computers can understand. APIGPT uses ChatGPT's natural language understanding ability, allowing users to interact with APIGPT in the same way they would with another human.

API designers can directly use APIGPT to train their own assistants, which can then be used through the command line or API. APIGPT also provides integration with the LanyingIM SDK, which allows developers to embed a chatbox in their API documentation as a learning assistant. Please refer to the APIGPT Usage Documentation for specific usage instructions.

We would like to thank the following open-source projects and teams:

• ChatGPT API by OpenAI - Provides natural language understanding technology.
• LanyingIM - Provides professional IMSDK.

If you would like to contribute code or report bugs to APIGPT, please raise an issue on GitHub.

APIGPT uses the MIT License.