DevChat is an open-source tool that helps you write prompts to generate code and documentation. It is not limited to performing predefined operations on short code snippets.

DevChat is the way developers interact and collaborate with AI.

💬 Build the future with prompts, not code.

Why DevChat?

To enable AI to write code, DevChat assists you in providing the right context for AI.

Chat history, code, files, directory trees, git diff --cached, or the output of any command.
To apply AI-generated code, DevChat streamlines your actions to take.

View diffs, copy or insert, commit & sync, or export to documentation, wikis, and more.
To guide AI in your work, define your own workflows with DevChat.

Prompt templates, iterative calls to AI, and program operations.
To suit your preferences, customize your own experiences with DevChat.

Open-source, no waiting for opaque feature schedules. Access GPT-4 today and more models in the future.

What is Prompt-Centric Software Development (PSSD)?

The traditional code-centric paradigm is evolving.
Write prompts to create code. Transform prompts into everything.
It is DevPromptOps

Installation

Open Visual Studio Code
Press Ctrl+P to open the Quick Open dialog
Type ext install merico.devchat and press Enter
Install devchat package by pip install devchat
setting your OPENAI_API_KEY in setting/DevChat

Usage

Open the chat panel by clicking on the "Chat with Bot" view in the Explorer or by running the "DevChat" command from the Command Palette (Ctrl+Shift+P).
Type your question or prompt in the input field and press Enter to send it to the AI bot.
To add code snippets or files to the chat, right-click on the selected code or file in the editor and choose "Add to DevChat" from the context menu.

Configuration

You can configure the following settings in your settings.json file:

DevChat.llmModel: Select the AI model to use (default: "OpenAI").
DevChat.maxLogCount: Limit the number of prompts to output (default: 20).
DevChat.OpenAI.model: Specify the AI model (default: "gpt-4").
DevChat.OpenAI.temperature: Specify the AI model temperature (default: 0.2).
DevChat.OpenAI.stream: Specify the AI model stream (default: true).
DevChat.OpenAI.tokensPerPrompt: Specify the number of tokens for each prompt (default: 6000).
DevChat.OpenAI.useHistoryPrompt: Use history prompts as context (default: true).
DevChat.OpenAI.apiKey: OpenAI API Key (default: "").
DevChat.OpenAI.EndPoint: OpenAI API server (default: "").

Contributing

If you have any suggestions or issues, please feel free to open an issue or submit a pull request on the GitHub repository.

Automated Publishing Process

Check out our Automated Publishing Process for a detailed walkthrough of how we manage the automated release of new versions for the DevChat VSCode Extension.

Contact Information

hello@merico.dev