vip-sin / appaiprompts

Github Prompts for Developers

Overview

This project focuses on prompt-based coding. This repo utilises Next.js web app development, python, javascript & .net snippets to understand the usage, leveraging AI assistants like GitHub Copilot and Tabnine to expedite code completion and enhance development productivity.

Objective

1. To utilize AI-powered tools to streamline the development process of Next.js web applications. By harnessing the capabilities of GitHub Copilot/Tabnine, developers aim to enhance code completion and reduce manual coding efforts.

1. What is Github Copilot?

• An AI-powered code completion tool developed by Github in collaboration with OpenAI.
• Utilizes machine learning models to assist developers in writing code by suggesting whole lines or blocks of code based on the context and code patterns.
• Iimprove developers' productivity by automating repetitive coding tasks and reducing the time spent on googling.
• Get intelligent code suggestions for various programming languages and frameworks, making it a versatile tool for a wide range of projects.

2. How to Use Github Copilot?

• Github Copilot subscription
• Install Github Copilot extension in your IDE e.g. VS Code.
• As you write code, Copilot suggests completions based on the code context and patterns.
• It provides suggestions for
  1. Entire lines of code,
  2. Function definitions,
  3. Variable names, and more. You can accept or reject these suggestions by simply pressing the Tab key or using other keyboard shortcuts.

Example

How prompts work to build logic based on a command or specification document by using the following prompts:

• Write a function that...: It generates a function that performs a specific task, like retrieving data from a database or sorting a list of numbers.

• Given the following command...: It generates code that executes a specific command, such as creating a new file or listing the contents of a directory.

• According to the following specification document...: It can be used to generate code that implements a specific set of requirements, such as a user interface or a data model.

• • Write a function that retrieves the user's name from the database.
  • Given the following command: ls -l, write code that lists the contents of the current directory.
  • According to the following specification document, write code that implements a user interface for a calculator.

It can help developers save time and improve their productivity. By using the prompts described above, developers can easily generate code that implements the logic they need.

• Additional tips to build logic based on a command or specification document:

  • Be as specific as possible in your prompts. The more specific you are, the better GitHub Copilot will be able to generate code that meets your needs.
  • Use natural language. GitHub Copilot can understand natural language, so you don't need to use any special syntax or keywords in your prompts.
  • Be patient. GitHub Copilot is still under development, so it may not always be able to generate the perfect code. However, it will get better over time, and it can still be a valuable tool for saving time and improving productivity.

How Copilot can assist with code completion:

    ## Example 1: Function Definition
    def calculate_average(numbers):
        """
        Calculates the average of a list of numbers.
        """
        total = sum(numbers)
        avg = total / len(numbers)
        return avg

    ## Example 2: Code Block
    if condition:
        # Copilot suggests completing the code block based on the condition
        do_something()
        do_another_thing()
        # ...

3. Free/Paid Versions

Github Copilot subscription-based pricing model.

• Enterprise - $19/user/Month
• Personal - $10/Month.
• Personal - $100/Year.

Tabnine subscription-based pricing model.

• Enterprise : TBD
• Basic Personal - $0/Month
• Pro Personal - $12/Month

4. Limitations/Risks/Vulnerabilities

While Github Copilot is a powerful tool, it does come with certain limitations and potential risks:

• AI Pair Programmer: It assits in coding but does not create something new.

• Code Sharing: Local, hybrid, Cloud : Its upto the developers for trusting the assistant on privacy

• Quality of Generated Code: The code suggestions provided by Copilot are based on patterns it has learned from vast code repositories. However, the quality of the generated code may vary, and manual review is essential to ensure correctness and adherence to project requirements.

• License and Legal Considerations: Copilot may generate code snippets that include copyrighted or licensed code. It is crucial for developers to review and modify the generated code to comply with applicable licenses and legal requirements.

• Security Vulnerabilities: Copilot's suggestions are based on patterns in existing code, which may include security vulnerabilities or insecure practices. Developers should exercise caution and perform proper security testing and code review to mitigate any potential vulnerabilities.

• Data Privacy: Take precautions to protect sensitive code data from unauthorized access or exposure. Be cautious when sharing code snippets that may contain proprietary or sensitive information.

• Code Ownership: Ensure that you have the necessary rights and permissions to use and share code snippets generated by GitHub Copilot. Respect intellectual property rights and avoid sharing code that may infringe upon copyrights or licenses.

• Code Review and Validation: While GitHub Copilot can provide helpful code suggestions, it is essential to manually review and validate the generated code for correctness, security vulnerabilities, and adherence to coding standards.

• Avoiding Sensitive Information: GitHub Copilot may learn from existing code repositories, including publicly available code. Be cautious not to input or rely on code snippets that contain sensitive information such as credentials, access tokens, or personally identifiable information (PII).

• Secure Development Environment: Ensure that your development environment, including the IDE and the system it runs on, is appropriately secured. Keep your tools and dependencies up to date to mitigate potential security vulnerabilities.

• Third-Party Integration Security: If GitHub Copilot integrates with other services or APIs, consider the security measures in place for those integrations. Evaluate their data handling practices and ensure that they align with your privacy and security requirements.

• Data Transmission: When using GitHub Copilot, be mindful of how code snippets are transmitted between your local environment and any remote services. Utilize secure communication protocols (e.g., HTTPS) to protect code snippets during transmission.

• User Awareness and Training: Educate developers and users about the privacy and security considerations associated with using GitHub Copilot. Promote awareness of best practices for handling sensitive code data and encourage adherence to security guidelines.

5. Advantages

Github Copilot offers several advantages to developers, including:

• Logic and Syntax Predictions : Code completion, logic analysis, alternatives and debugging are all available with github copilot.

• Unit Test Cases : Code completion, logic analysis, alternatives and debugging are all available with github copilot.

• Improved Productivity: Copilot speeds up the coding process by providing accurate code suggestions, reducing the time spent on repetitive tasks and searching for examples.

• Learning and Knowledge Transfer: Copilot can introduce developers to new coding patterns, best practices, and libraries through its generated suggestions, enabling learning and knowledge transfer within the development team.

• Reduced Cognitive Load: By automating code completion, Copilot helps reduce the cognitive load on developers, allowing them to focus more on the logic and higher-level aspects of their code.

Name	Id	Description	Version	Publisher	Marketplace
Tabnine AI Autocomplete for Javascript, Python, Typescript, PHP, Go, Java, Ruby & more	TabNine.tabnine-vscode	JavaScript, Python, Java, Typescript & all other languages - AI Code completion plugin. Tabnine makes developers more productive by auto-completing their code.	3.6.56	TabNine	Link
GitHub Copilot	GitHub.copilot	Your AI pair programmer	1.89.156	GitHub	Link
Blackbox AI Code Generation, Code Chat, Code Search	Blackboxapp.blackbox	Code as fast as you think	1.0.21	BLACKBOX AI	Link
Codeium: AI Coding Autocomplete and Chat for Python, Javascript, Typescript, Java, Go, and more	Codeium.codeium	The modern coding superpower: free AI code acceleration plugin for your favorite languages. Type less. Code more. Ship faster.	1.2.36	Codeium	Link
Cody AI by Sourcegraph	sourcegraph.cody-ai	AI code assistant that writes code and answers questions for you	0.2.3	Sourcegraph	Link
CodeComplete	CodeComplete.codecomplete-vscode		0.0.8	CodeComplete	Link
AWS Toolkit (Includes Codewhisperer)	AmazonWebServices.aws-toolkit-vscode	Including support for CodeWhisperer, CodeCatalyst, Lambda, S3, CloudWatch Logs, and many other services	1.77.0	Amazon Web Services	Link
Fauxpilot	Venthe.fauxpilot	Get completions from Fauxpilot server	1.1.5	Venthe	Link
Tabby	TabbyML.vscode-tabby	Get completions from Tabby server	0.0.6	TabbyML	Link

transform data from CSV format to JSON/XML format using a specified logic built upon the command/specification document.

Note

ChatGPT uses your input via the web interface to train its model. This is how Samsung employees leaked confidential data by asking ChatGPT to generate meeting notes. ChatGPT retains user data, even that of paying users. If you don’t consent to this, you need to opt out.

Comparision(https://blog.pragmaticengineer.com/github-copilot-alternatives/)

Yet to explore

• Logic build up using command/specification document
• Format Transformation from csv to JSON/XML
• Logging and exception handling with command on the code