Content
[English](README_EN.md) | **English**
# GreenCandle v0.2.0
## Project Introduction
- GreenCandle (official name) is a text creation tool that primarily explores the application of AI in text creation.
- Available platforms: ubuntu24.04, windows11
- Respecting different users' habits, no pre-set "one-click functions" are provided.
- The goal is to integrate AI into the writing process, enhancing efficiency and extending the author's thoughts, rather than replacing the author.
# Core Features
## 0. Built-in Butler Agent
Allows AI to operate backend APIs directly, replacing user operations (currently in the early stages).
## 1. Tool Invocation
Consistent with core logic of cursor/cline/roo code/opencode/kilo code, with some design differences.
Supports human-in-the-loop, encouraging frequent user intervention in AI content generation and timely corrections.
Also allows automatic approval, enabling AI to execute tasks completely autonomously.
## 2. Themes
Supports day/night modes and custom theme coloring.
## 3. RAG Knowledge Base
Allows AI to retrieve necessary information, enhancing effects.
- Two-step RAG: Refers to retrieving relevant information and returning it to AI along with other context before calling AI.
- Agentic RAG: Empowers AI with two knowledge base tools, allowing it to decide whether to query data.
## 4. General Search
Standard feature of text editors.
Allows users to search text.
## 5. Checkpoint
Standard feature of text editors.
Allows users to archive, restore, and view differences between file content checkpoints.
## 6. Provider Adaptation
Supports 11 providers.
Supports custom providers (openai compatible).
Includes an embedded model, ready to use.
## 7. MCP Client
Supports connecting to MCP servers.
## 8. Skills Support
Can access various skills, similar to openclaw and claude code.
## 9. Others:
* **Command Execution**: Allows AI to execute commands directly, such as python, npm, etc.
* **Terminal**: Similar to terminals in editors like vscode, allows command line execution.
......
# Tech Stack
## Frontend Technology
## Backend Technology
## Development and Build Tools
Total : 215 files, 18327 codes, 3673 comments, 3443 blanks, all 25443 lines
## Languages
| language | files | code | comment | blank | total |
| :--- | ---: | ---: | ---: | ---: | ---: |
| TypeScript JSX | 76 | 8,329 | 627 | 1,022 | 9,978 |
| Python | 63 | 4,458 | 2,210 | 1,349 | 8,017 |
| TypeScript | 38 | 3,060 | 632 | 587 | 4,279 |
| Markdown | 11 | 946 | 0 | 276 | 1,222 |
| JavaScript | 10 | 489 | 91 | 91 | 671 |
| YAML | 3 | 470 | 59 | 37 | 566 |
| PostCSS | 4 | 185 | 19 | 30 | 234 |
| JSON | 3 | 167 | 0 | 2 | 169 |
| PowerShell | 1 | 64 | 11 | 15 | 90 |
| Skill | 2 | 62 | 0 | 21 | 83 |
| pip requirements | 1 | 34 | 0 | 0 | 34 |
| JSON with Comments | 1 | 29 | 19 | 7 | 55 |
| Ignore | 1 | 20 | 5 | 5 | 30 |
| HTML | 1 | 14 | 0 | 1 | 15 |
## Quick Start
### Installation & Launch
1. **Clone Repository**:
```bash
git clone git@github.com:FlickeringLamp/ai-novelist.git
cd ai-novelist
```
2. **Install Frontend Dependencies**:
Enter the frontend directory (`frontend/`) and install dependencies, build frontend, and start:
```bash
cd frontend
npm install
npm run build
npm start
```
3. **Install Backend Dependencies**:
Create a virtual environment from the root directory (`ai-novelist`), activate it, and install backend dependencies, then start:
```bash
python -m venv backend_env
backend_env\Scripts\activate
cd backend
pip install -r requirements.lock
cd ..
python main.py
```
> **Note**: Using `requirements.lock` ensures all dependency versions are completely locked, avoiding compatibility issues. For development environments requiring the latest versions, use `requirements.txt`.
4. **Browser Access**:
Access http://localhost:3000 in your browser.
5. **Other Launch Methods**:
From the root directory:
```bash
cd frontend
npm run electron-dev
```
**Note**:
When launching with electron-dev, terminal functionality is available, but not on the web due to Node.js main process limitations.
## Contribution
We welcome contributions in all forms! If you find bugs, have feature suggestions, or want to submit code, please participate through GitHub Issues or Pull Requests.
For detailed guidelines, see:
- [Contribution Guide](CONTRIBUTING.md) - How to submit Issues, PRs, and other workflows.
- [Development Specifications](DEVELOPMENT.md) - Code style, project structure, and other specifications.
We do not oppose AI-assisted programming, but ensure your agent understands this file's requirements and that you understand what AI is writing; otherwise, it may not be merged.
- [Agent Code of Conduct](AGENT.md)
To maintain the project's healthy development, ensure:
- Submitted code is compatible with the [MIT License](LICENSE).
- Follow conventions in [Development Specifications](DEVELOPMENT.md).
Thanks for the support of every contributor!
## License
This project uses the [MIT License](LICENSE).
## Acknowledgements
The development of this project draws on `opencode` and `roo-code` projects to some extent. We express our sincere gratitude to the developers of these projects.
- The `opencode` project is open-sourced under the MIT License; see [`LICENSE-opencode.txt`](./LICENSE-opencode.txt).
- The `roo-code` project is open-sourced under the Apache License 2.0; see [`LICENSE-roo-code.txt`](./LICENSE-roo-code.txt).
