Easy-to-use open source AI writing assistant

General Introduction

AI Writing Assistant is an open-source AI writing assistant based on Next.js, hosted on GitHub, designed to help writers, content creators, and copywriting professionals improve the efficiency and quality of their writing through artificial intelligence technology. The tool allows users to organize their writing needs with a detailed prompt style editor and send commands to multiple Large Language Models (LLMs) to generate content. It supports major API interfaces such as OpenAI, Claude, and Grok, and offers flexible configuration options and a user-friendly interface. Whether it's creating articles, adjusting tone, or exporting Markdown files, this project provides users with efficient writing support and is an excellent example of how technology and creativity can work together.

Function List

• Cue word style customization: Support multi-dimensional editing such as language, structure, emotion, etc., to generate text that meets the user's needs.
• Multi-model supportIt is compatible with OpenAI, Claude, Gemini and other LLM APIs to meet the needs of different scenarios.
• Flexible API configuration: Easily set up API keys and endpoints with a collapsible panel.
• Real-time content editing: The generated text can be modified directly in the built-in editor.
• Markdown export: One-click export of content to Markdown format for easy saving and sharing.
• Debugging Information View: Provide detailed API response logs for troubleshooting.
• Theme switching: Supports dark and light color modes to adapt to user preferences.
• Agent Support: Built-in proxy functionality to address network access or CORS limitations.

Using Help

AI Writing Assistant is an open source tool that requires users to get the code via GitHub and run it locally. Below is a detailed installation and usage guide to help users get started quickly.

Installation process

To use this tool, you'll need some technical foundation, including a Node.js environment and Git operations. Here are the steps:

1. Preparing the environment
   • Ensure that your computer has Node.js 16.20.0 or later installed (Download:https://nodejs.org/).
   • Install the Git tool (download address:https://git-scm.com/) for cloning code.
2. Get Project Code
   • Open your browser and visit https://github.com/GeekyWizKid/writing-helperThe
   • Click on the green "Code" button to copy the repository address:https://github.com/GeekyWizKid/writing-helper.gitThe
   • Open a terminal and enter the following command to clone the project:

     git clone https://github.com/GeekyWizKid/writing-helper.git
     

   • Go to the project catalog:

     cd writing-helper
     

3. Installation of dependencies
   • Run the following command in the terminal to install the required dependencies for the project:

     npm install
     

   • Wait a few minutes until all dependencies are installed.
4. Starting the Development Server
   • Enter the following command to start the project:

     npm run dev
     

   • After a successful launch, open your browser and visit http://localhost:3000You can see the tool interface.
5. troubleshooting
   • If the startup fails, check that the Node.js version is correct, or see the terminal error message.
   • Visit GitHub's "Issues" page to search for similar issues or submit a new issue for help.

Main Functions

After installation, users can use the following functions through the browser interface:

API Settings

• Open the settings panel: Once on the homepage, click the "API Settings" button on the left to expand the configuration panel.
• Select Model: Select a supported API (e.g. OpenAI, Claude, Grok, etc.) in the drop-down menu.
• Enter the key: Enter your API key (obtained from the appropriate platform, e.g., the OpenAI website) in the corresponding field.
• Custom Endpoints: If you are using a proxy or local model, you can enter a custom address in the "API Base URL".
• Save Configuration: Click "Save" or go directly to the next step and the configuration will take effect automatically.

Content setup and style customization

• Setting the theme: Fill in the "Article Topic" input box with the goal of the creation, e.g. "Scientific Press Release".
• Add Keywords: Enter relevant words in the "Keywords" field, such as "Artificial Intelligence, Technology Trends", separated by commas.
• Target word count: Enter the desired number of words, for example 500.
• Style Editor::
  1. Click the "Style Editor" button to enter the detailed settings.
  2. style: Enter an overall description, e.g. "concise and clear".
  3. language: Choice of sentence types (short sentences predominate) or rhetoric (e.g., similes).
  4. framework: Set the length of the paragraph (3-5 sentences per paragraph) and the transitions (natural segue).
  5. move (emotionally): Adjust the tone (e.g., positive optimism).
  6. Save and return to the main screen.

Generate and Edit Content

• Generate content: Click the "Generate Content" button, wait a few seconds and the right editor will show the result.
• Real-time modification: Adjust the text directly in the editor, for example by removing redundant sentences or adding details.
• Exporting files: Click the "Export as Markdown" button, select the save path, the file will be saved as .md Format Download.

Debugging and Optimization

• View Log: If the generation fails, click "Debug Info" to view the API response details and check for key or network issues.
• Using Proxies: If you encounter CORS error, check "Use proxy server" in API settings to solve the problem by built-in proxy.

Featured Functions

• Multi-model switching::
  Users can switch models at any time, for example, from OpenAI GPT-4 to Grok, to compare the generation effect of different models. Steps: Enter API settings, select new model, enter corresponding key, save and regenerate.
• In-depth customization of styles::
  The style editor is the central highlight. For example, set the narrative point of view to "first person," the emotional tone to "humor and lightheartedness," and the pacing to "short, fast-paced sentences," and the generated content will strictly follow these requirements.
• Grok API Testing::
  If using Grok Models, accessible http://localhost:3000/grok path, test the connection status and view detailed response logs.

Recommendations for use

• initial use: We recommend trying OpenAI's GPT-3.5-Turbo model first, which is easy to configure and responsive.
• Save the key: To avoid duplicate entries, it is recommended that the API key be recorded in a secure location.
• Community involvement: If the functionality is not perfect, you can contribute code or suggestions by submitting a Pull Request on GitHub.

With these steps, users can take full advantage of the powerful features of the AI Writing Assistant, whether it is to quickly generate drafts or deeply customize content, which can significantly improve writing efficiency.