API Token Manager
A VS Code extension to securely store and manage API tokens for multiple AI providers. Switch between providers and models seamlessly.
Features
- Secure Storage: Tokens stored in OS keychain via VS Code's
SecretStorage — never in settings or disk
- Multi-Provider Support: Built-in configs for OpenAI, Anthropic, Google (Gemini), Groq, Ollama, OpenRouter
- Custom Providers: Add your own providers with custom base URLs, auth headers, and models
- Model Selection: Switch models per provider with metadata (context window, vision, tools)
- Connection Testing: Verify tokens work with one click
- Zero Config: Works out of the box with sensible defaults
Installation
From VSIX (Recommended)
# Build the extension
npm run compile
vsce package
# Install in VS Code
code --install-extension api-token-manager-0.1.0.vsix
From Source
git clone https://github.com/your-username/api-token-manager
cd api-token-manager
npm install
npm run compile
# Press F5 in VS Code to launch Extension Development Host
Quick Start
- Open Command Palette (
Ctrl+Shift+P / Cmd+Shift+P)
- Run
API Token Manager: Set API Token
- Select your provider (e.g., OpenAI)
- Paste your API key
- Run
API Token Manager: Test API Connection to verify
Commands
| Command |
Description |
Set API Token |
Store a token for a provider (secure input) |
Clear API Token |
Remove a stored token |
Test API Connection |
Verify token works with provider's API |
Select Provider |
Switch active provider (OpenAI, Anthropic, etc.) |
Select Model |
Switch active model for current provider |
Configure Provider |
Add/edit/remove custom providers |
Supported Providers
| Provider |
Models |
Vision |
Tools |
Notes |
| OpenAI |
GPT-4o, GPT-4o-mini, GPT-4, GPT-3.5 |
✅ |
✅ |
|
| Anthropic |
Claude 3.5 Sonnet, Haiku, Opus |
✅ |
✅ |
|
| Google |
Gemini 1.5 Pro, Flash, 1.0 Pro |
✅ |
✅ |
|
| Groq |
Llama 3.1 70B/8B, Mixtral, Gemma |
❌ |
✅ |
Fast inference |
| Ollama |
Llama 3.1/3.2, Mistral, CodeLlama |
❌ |
❌ |
Local (localhost:11434) |
| OpenRouter |
100+ models via single API |
✅ |
✅ |
Aggregator |
Custom Providers
Add any OpenAI-compatible API:
- Run
Configure Provider → Add Custom Provider
- Fill in:
- ID: Unique identifier (e.g.,
my-company)
- Name: Display name
- Base URL:
https://api.mycompany.com/v1
- Auth Header:
Authorization: Bearer {token} (or custom)
- Test Endpoint:
/models (or /health)
- Models: Comma-separated
id:Name pairs
- Set token via
Set API Token
Example for Azure OpenAI:
Base URL: https://my-resource.openai.azure.com/openai/deployments/my-deployment
Auth Header: api-key: {token}
Test Endpoint: /chat/completions?api-version=2024-02-15
Architecture
src/
├── config/
│ └── providers.ts # Built-in provider definitions
├── services/
│ ├── tokenManager.ts # SecretStorage wrapper (per-provider)
│ └── providerManager.ts # Active provider/model state (globalState)
├── api/
│ └── clientFactory.ts # Provider-specific API clients
├── commands/
│ ├── setToken.ts
│ ├── clearToken.ts
│ ├── testConnection.ts
│ ├── selectProvider.ts
│ ├── selectModel.ts
│ └── configureProvider.ts
└── extension.ts # Activation entry point
Security
- Tokens encrypted in OS keychain (Windows Credential Manager, macOS Keychain, Linux Secret Service)
- Never written to
settings.json, globalState, or logs
- Input boxes use
password: true masking
- No telemetry by default
Settings
| Setting |
Default |
Description |
api-token-manager.defaultProvider |
openai |
Provider on first run |
api-token-manager.requestTimeout |
60000 |
Request timeout (ms) |
api-token-manager.enableTelemetry |
false |
Anonymous usage stats |
Development
npm install # Install dependencies
npm run compile # Compile TypeScript
npm run watch # Watch mode
npm run lint # ESLint
npm run package # Create .vsix
License
MIT — see LICENSE
Contributing
- Fork the repo
- Create a feature branch
- Add tests for new providers
- Submit a PR
Built for developers who switch between AI providers daily.
