GitGeminiHub

About The Project

This is GitGeminiHub built by GitHub Action and Google Gemini API to response automatically by submitting Issues.

It is a learning, open-source project. If any complains, please contact me. If you are interested in this project，it would be honored to receive yours Star and Fork

Built With

• GitHub Action: Automate your workflow from idea to production
• Google Gemini: Welcome to the Gemini era

Getting Started

Follow the steps learn how to submit a task by open issue

Prerequisites

For developers or contributors who folk as a personal project:

1. Fork repository as your owners
2. Get a Gemini API key
3. Add the Gemini API key in project setting
4. Settting -> Security/Secrets and variables/Actions -> Secrets/Repository secrets

For users:

1. You need a GitHub account and should better know how to use GitHub, like create an issue.
2. Open Issue to create a New issue ticket
3. Choose an issue template to submit a request
4. Submit the issue ticket

Steps

1. Open Issue to create a New issue ticket
2. Choose any task request to start submitting a task issue
3. Fill the prompt you want task to execute in ### Prompt section
4. Fill the content (upload/drop files or images or their links if needed) in ### Content section
5. Submit the new issue
6. Wait the task finished.
7. Check the new comment added by action bot. If task failed, please check the issue description, edit or reopen or re-create can re-trigger the translation task automatically.

template explanation

When you try to submit the ticket, you will see the pate like below:

1. Please just fill the below task information as follows and DO NOT remove any text > of this Description template
2. Fill your prompt in the ### Prompt section between line marks ---
3. Fill additional image or text files in the ### Content section between line marks ---
4. If the task fails, please check the Safety settings and adjust the settings in the json data in ### Setting section

Let's read images

See Introduction for details.

Prompt

---

Please recognize the text in the picture and convert to markdown

---

Content

---


---

Introduction

balabalabalabalabalabalabalabalabalabalabalabalabalabalabalabalabala

Setting

Here is the generation configuration and safety setting about Gemini, you can modify them according to your needs. Safety settings

---

  {
    "model_name": "gemini-pro-vision",
    "generation_configuration":
    {
      "temperature": 0.9,
      "top_p": 1.0,
      "top_k": 1,
      "max_output_tokens": 2048
    },
    "safety_setting":
    {
      "harassment": "BLOCK_MEDIUM_AND_ABOVE",
      "hate_speech": "BLOCK_MEDIUM_AND_ABOVE",
      "sexually_explicit": "BLOCK_MEDIUM_AND_ABOVE",
      "dangerous_content": "BLOCK_MEDIUM_AND_ABOVE"
    }
  }

---

Supported task templates

• Task - Ask: Recommand ask Gemini for anything you want, like google or bing anything - writing, ask question, translate or solve math problems.
• Task - Ask with images: it can read images and do something as your description - image apprehension, text extraction and others.
  • Examples:
  • What's in the picture?
  • Write a short, engaging blog post based on this picture. It should include a description of the meal in the photo and talk about my journey meal prepping.
  • Please recognize the text in the picture and convert to markdown
  • Describe what's in each image in detail and summary the common between these
• Task - Ask with file: It can read a file if you attach, and post the result one to you based on your prompt.
  • Examples:
  • Please summary the text: ..balabala.. .
  • Please polish the text: ..balabala.. .
  • Please translate the text: ..balabala.. .
• More task templatesa are in support ... If you have more expected features, please choose Feature requst ticket to ask for new features.

Examples

Here are some examples you may be interested in. If you have awesome tasks/examples by this project, please submit a pull request or @xwnb. I'll list them below.

1. Example: Add contextual information
2. Example: Summarize this text
3. Example: Translate the attached to Chinese
4. Example: Parse the formula
5. Example: Describe what's in each image in detail and summary the common between these?
6. Example: Parse the table in this image into markdown format
7. Example: Create an outline for an essay about hummingbirds
8. Example: Recognize the text in the picture and convert to markdown
9. Example: Please recognize the text in the picture and translate to Chinese
10. Example: Translate pdf to Chinese [NEW 2024/6/8]

Roadmap

• [x] Build with GitHub Action and Gemini API triggered by Issue event
• [x] Support .txt, .md files for translation
• [x] Support image file for prompting
• [x] Post response by Issue comment.
• [x] Add configuration section for Gemini setting in Issue template
• [ ] Support other plain text formats
• [x] Support *.pdf phrase [NEW 2024/6/8]
• [ ] Support OCR for images, scanned-documents
• [ ] Notify task result to users by email
• [ ] ...

See the [open issues]() for a full list of proposed features (and known issues).

Contributing

Contributions are what make the open source community such an amazing place to learn, inspire, and create. Any contributions you make are greatly appreciated.

If you have a suggestion that would make this better, please fork the repo and create a pull request. You can also simply open an issue with the tag "enhancement". Don't forget to give the project a star! Thanks again!

1. Fork the Project
2. Create your Feature Branch (git checkout -b feature/AmazingFeature)
3. Commit your Changes (git commit -m 'Add some AmazingFeature')
4. Push to the Branch (git push origin feature/AmazingFeature)
5. Open a Pull Request

License

Distributed under the MIT License. See LICENSE.txt for more information.

Contact

Project Link: https://github.com/xwnb/GitGeminiHub

Acknowledgments

• GitHub Action: Automate your workflow from idea to production
• Google Gemini: Welcome to the Gemini era

(back to top)