Skip to content

TabbyML/quick-question

Repository files navigation

⁉️ QuickQuestion

License Docker build status

🤔 What is this?

An incubating AI-powered Q&A for your codebase.

Live Demo

Example Quick Question

🚀 Deployment

Make sure git-lfs is installed.

  1. Clone the repository.

    git clone https://github.com/TabbyML/quick-question.git
  2. Save your OPENAI_API_KEY to file as secret.

    echo YOUR_OPENAI_API_KEY > openai_api_key.txt
  3. Optional Run Quick Question on your github project.

    1. Create new directory for you project under /data, e.g /data/quick-question.

    2. Add a new metadata.json file in your project directory. Here is a templete of file content, replace {GITHUB_PROJECT} with your own project name, e.g TabbyML/quick-question.

      {
          "name": "{GITHUB_PROJECT}",
          "exampleQueries": ["How to ...?"]
      }

      See ./data/diffusers/metadata.json for a complete example.

  4. Start container.

    docker-compose up
    

🛠️ Development

  1. Make sure git-lfs is installed.
  2. Clone the repository, runs yarn to install dependencies.
  3. Run yarn lerna run build.
  4. Switch workdir to ./packages/quick-question.
  5. Copy .env.sample to .env.local, and set your OPENAI_API_KEY.
  6. Run yarn dev to start local development

❤️ Acknowledgement

Many thanks to WizAI for contributing with code-search, a project that QuickQuestion branched from.