GearCodeAI — AI for Robotics

AI-powered coding assistant for ROS 2, embedded C++, and drone development. Phase 1 MVP — VS Code Extension

---

What it does

Feature	How to trigger
Ask AI about selected code	Highlight → Ctrl+M (or Cmd+M on Mac)
Auto-detect build errors	Just run colcon build — GearCodeAI watches the terminal
Fix build error with AI	Click the status bar item or run GearCodeAI: Fix Last Build Error
Open AI chat panel	Command Palette → GearCodeAI: Open AI Chat
Show workspace context	Command Palette → GearCodeAI: Show Workspace Context

---

Setup

1. Get a free NVIDIA NIM API key

1. Go to build.nvidia.com
2. Sign up / log in → click Get API Key
3. Copy the key (starts with nvapi-)

2. Set your key in VS Code

Open Settings → search gearcodeai.nimApiKey → paste your key.

Or via the command palette:

Preferences: Open User Settings (JSON)

Add:

"gearcodeai.nimApiKey": "nvapi-your-key-here"

3. Open your ROS 2 workspace

Open the folder that contains your src/ and (optionally) colcon.meta. GearCodeAI will auto-detect your packages on startup and inject them as context into every AI query.

---

How it knows your project

On activation, GearCodeAI scans for:

• package.xml — package name, build type (ament_cmake / ament_python), dependencies
• CMakeLists.txt — add_executable() targets (node names)
• setup.py — Python node entry points
• Topic name string literals in src/ files

This context is injected into the AI system prompt so answers are specific to your workspace.

---

Configuration

Setting	Default	Description
gearcodeai.nimApiKey	""	NVIDIA NIM API key
gearcodeai.nimModel	meta/llama-3.3-70b-instruct	Model to use
gearcodeai.errorSnifferEnabled	true	Watch terminals for build errors
gearcodeai.autoAnalyzeWorkspace	true	Analyze on startup

---

Build from source

cd GearCodeAI
npm install
npm run compile        # or: npm run watch
# Press F5 in VS Code to launch Extension Development Host

To package a .vsix:

npm run package
# → gearcodeai-0.1.0.vsix

---

Roadmap

• Phase 2 — One-click Wi-Fi deploy to RPi/Jetson, Docker cross-compilation, .bag/ulog telemetry reader
• Phase 3 — VS Code Marketplace launch, user metrics, community feedback
• Phase 4 — Native standalone IDE (VSCodeium fork), live sensor graphs, 3D physics simulator

---

License

MIT — see LICENSE.