Wingman

Meet Wingman — the AI that assists, not insists. Wingman is your sidekick that debugs, explains, and tests… only when summoned.
Forget copy-pasting into ChatGPT or losing focus jumping between tabs. Wingman lives in your editor, ready to break down errors, suggest smart fixes, or whip up test cases — without taking the wheel.

Because real devs don’t want AI writing their code — they want AI helping them write better code.

✨ Features

1. Code Walkthrough

Understand what a piece of code is doing, step by step. Select the code you want explained — Wingman reads your selection and breaks it down in plain language, helping you grasp logic, flow, and potential pitfalls.

2. Generate Testcases

Wingman generates sample test cases for your functions or selected code — right in your editor.
Support for edge cases and property-based testing is coming soon.

3. Explain Errors

Wingman reads your terminal output, stack traces, or compile-time errors and explains what went wrong — in English.

4. Suggest Fixes

Wingman analyzes your code and its errors to suggest contextual, intelligent fixes — and teaches you why it works.

⚙️ Setup & Configuration

Wingman supports multiple LLM providers. You can choose the one that works best for you:

1. Select Your LLM Provider

For now we only have Google. Will bring others and Ollama in future versions.

2. Choose the Model

Pick your preferred model like gemini-1.5-pro, gemini-1.5-flash, etc. Wingman lets you switch models anytime based on your needs and budget.

3. Set Your API Key

On first run (or when using a new provider), Wingman will prompt you to enter your API key.
These keys are stored securely using VS Code’s Secrets API.

You can set/change them from the command palette:
Wingman: Set API Key

4. Reset All Secrets

Want a fresh start? Use the command:
Wingman: Reset Secrets
This will clear all stored keys and provider/model settings.

🔧 Requirements

None yet — Wingman runs out of the box!

Enjoy. 🛩️