AI-Powered API Testing Desktop Application
Apilot is a desktop application for API testing and validation powered by AI. Import OpenAPI/Swagger specifications, generate intelligent test cases using AI services, and execute tests with comprehensive assertions and environment management.
- 📋 Spec Management: Import and manage OpenAPI 3.x and Swagger 2.0 specifications
- 🤖 AI-Powered Test Generation: Automatically generate comprehensive test cases using OpenAI, Anthropic, Google Gemini, or local Ollama models
- 🧪 Test Execution: Execute single tests or multi-step workflow tests with real-time results
- 🌍 Environment Management: Manage multiple environments with variable substitution per specification
- ✅ Assertions: Define and validate response assertions with JSONPath support
- 💾 Local Storage: All data stored locally using IndexedDB (no cloud/server required)
- 🖥️ Desktop App: Built with Electron for cross-platform support (macOS, Windows, Linux)
- Node.js 18+
- npm
npm installnpm run dev# Build for all platforms
npm run build
# Platform-specific builds
npm run build:mac
npm run build:win
npm run build:linux- Frontend: React 19 + TypeScript + TailwindCSS
- Desktop Framework: Electron 38
- Database: IndexedDB (via Dexie.js)
- State Management: TanStack Query (React Query)
- UI Components: Radix UI
- AI Services: OpenAI, Anthropic Claude, Google Gemini, Ollama
- Build Tool: Vite
apilot-app/
├── electron/ # Electron main process
│ ├── main.ts # Main process entry point
│ └── preload.ts # Context bridge (IPC)
├── src/
│ ├── pages/ # Application pages
│ │ ├── Specs.tsx # API Specifications list
│ │ ├── SpecDetail.tsx # Spec details & test generation
│ │ ├── Tests.tsx # Test cases management
│ │ └── Settings.tsx # Application settings
│ ├── components/ # Reusable React components
│ ├── lib/
│ │ ├── ai/ # AI service integrations
│ │ ├── api/ # IndexedDB operations (Dexie)
│ │ ├── converters/ # OpenAPI/Postman converters
│ │ ├── executor/ # Test execution engine
│ │ ├── hooks/ # React Query hooks
│ │ └── db.ts # Database schema
│ └── types/ # TypeScript type definitions
└── docs/ # Detailed documentation
OpenAPI/Swagger JSON files that define your API endpoints, parameters, request/response schemas.
Individual test scenarios generated from specs or created manually:
- Single Tests: Test a single API endpoint
- Workflow Tests: Multi-step tests that chain multiple endpoints with variable extraction
Per-specification variable sets for different deployment stages:
- Base URLs
- Authentication tokens
- Dynamic test data
- Variable substitution using
{{variableName}}syntax
Validation rules that verify API responses:
- Status code checks
- Response time validation
- JSONPath field assertions
- Header verification
- Body content matching
- Import Spec → Upload OpenAPI/Swagger JSON file
- Generate Tests → Select endpoints and use AI to generate test cases
- Configure Environment → Set up per-spec variables
- Execute Tests → Run tests and view real-time results
- Refine → Edit tests, add assertions, adjust parameters
Supports multiple AI providers (configured in Settings):
- OpenAI (GPT-4, GPT-4 Turbo, GPT-3.5)
- Anthropic (Claude 3.5 Sonnet, Opus, Haiku)
- Google Gemini (2.0 Flash, 1.5 Pro)
- Ollama (Local models: Llama, Mistral, etc.)
Tests are generated based on:
- Endpoint specifications and schemas
- Request/response examples
- Common API testing patterns (positive & negative cases)
- Workflow analysis for multi-step scenarios
Contributions are welcome! Please feel free to submit issues or pull requests.
MIT License - see LICENSE file for details
For issues, questions, or contributions, please create an issue.
