Streamline Code Review, Commit & Speed Up Dev Process. Your Own Personal Senior Engineer For Free!
Table of Contents
AI DEV SCRIPTS, leverages Local LLMs to streamline code improvement workflows and enhance your coding. It includes scripts like ai_review for suggestion generation, ai_pr for pull request analysis, and ai_commit for suggested commit messages, ai_chat for full RAG compatible chat feature. Additionally, it features an ai_readme script that generates customized readmes based on directory locations. Overall, it utilizes Ollama's DeepSeek Coder, and Mistral model to automate code improvements, security checks, and documentation within the repository ecosystem.
- AI Review: Scours through specified file formats and requests AI-generated suggestions for improvement.
- AI PR: Analyzes GitHub Pull Requests by calling an external Ollama DeepSeek Coder service.
- AI Commit: Generates commit messages using an AI model, adhering to conventional commit style and active voice guidelines.
- AI Readme: Generates customized readmes based on the repository's location, utilizing OpenAI's Ollama API and the Mistral model.
- AI Chat: Chat with websites, pdf, and markdown files, a RAG in your own terminal.
- AI Describe: Describe an image.
| File | Summary |
|---|---|
| ai_review | This ai_review file initiates the script that scours through specified file formats and requests AI-generated suggestions for improvement. It generates a markdown file containing improvements, best practices, readability enhancements, maintainability tips, and potential code examples, creating an impactful code improvement workflow within this repository's ecosystem. |
| ai_pr | The ai_pr script analyzes GitHub Pull Requests by calling an external Ollama DeepSeek Coder service. It generates brief summaries and flags potential security or coding best practices issues from the presented git changes. This tool supports automated PR review processes in the given repository infrastructure. |
| ai_commit | The ai_commit script in this repository's scripts folder is designed to generate commit messages using an AI model. This tool runs the deepseek-coder model from Ollama to suggest a commit message based on the git diff provided as input, adhering to conventional commit style and active voice guidelines. |
| ai_readme | Generate readme files for directories using the AI, named ai_readme script. The script triggers an AI to produce customized readmes based on the repository's location, utilizing OpenAI's Ollama API and the Mistral model. Emojis and flat-square badge styles are incorporated in the readme generation process. |
| ai_chat | Chat with web pages, PDFs, or markdown files of any size. Complete rag functionality. |
| ai_describe | Describe an image (uses the llava model). |
| openai | General purpose openai script. It's a dependency for several other scripts here. |
Requirements:
- Bash
- readmeai
- ollama
- deepseek-coder
- mistral
1. Clone the repository:
git clone https://github.com/ikramhasan/AI-Dev-Scripts.git2. Install the required dependencies:
pip install readmeai3. Install ollama: Download the latest release from here
4. Install deepseek-coder:
ollama run deepseek-coder:6.7b-instruct5. Install mistral:
ollama run mistral:7b-instruct6. Make the scripts executable:
chmod +x ./ai_review7. (Optional) Add the scripts to your PATH:
export PATH=$PATH:/path/to/AI-Dev-ScriptsNavigate to the directory where you want to run the script and execute the command below:
$ ./ai_review file.py file.js # for specific filesOr,
$ ./ai_review *.py *.js # for all files with .py and .js extensions
Navigate to the directory where you want to generate the readme and run the command below:
$ ./ai_readme
Navigate to your repo and run ai_commit using the command below:
$ ./ai_commit
Copy the pr link and run the command below:
$ ./ai_pr <pr_link>
Navigate to the directory where you want to run the script and run the command below:
$ ./ai_chat -t md -f blog.md -q "What is this blog about?"Or,
$ ./ai_chat -t pdf -f blog.pdf -q "What is this blog about?"Or,
$ ./ai_chat -t web -f https://www.example.com -q "What is this blog about?"
Navigate to the directory of your image and run the command below:
$ ./ai_describeThen follow the on-screen instructions.
The script requires
jqto be installed. Run the following command to install it.brew install jqAfter jq is installed, add the OPENAI_API_KEY variable to your path by running this command:
export OPENAI_API_KEY=your_openai_api_keyThen run the script using the command below:
$ ./openai <prompt>Then follow the on-screen instructions.
Contributions are welcome! Here are several ways you can contribute:
- Report Issues: Submit bugs found or log feature requests for the
scriptsproject. - Submit Pull Requests: Review open PRs, and submit your own PRs.
This project is protected under the MIT License.