feat: add AI-powered command suggestions for unrecognized commands

AGENTS.md

@@ -1,275 +1 @@
-# Safe CLI Development Guide
-
-## 🚨 CRITICAL SAFETY WARNING 🚨
-
-**NEVER run tests without isolated storage!** Integration tests were previously written in a dangerous way that could **DELETE YOUR ACTUAL WALLET DATA AND SAFE CONFIGURATIONS**.
-
-### Mandatory Safety Rules:
-
-1. **ALL integration tests MUST use `createTestStorage()`** from `src/tests/helpers/test-storage.ts`
-2. **NEVER instantiate storage classes without the `cwd` option** in test mode
-3. **ALWAYS verify tests are using `/tmp` directories** before running
-4. **Backup your config before running tests** if unsure
-
-The storage classes now have built-in safety checks that will throw an error if you try to use non-temp directories in test mode.
-
-### Safe Test Pattern (REQUIRED):
-
-```typescript
-import { createTestStorage } from '../helpers/test-storage.js'
-import { WalletStorageService } from '../../storage/wallet-store.js'
-
-describe('My Test', () => {
-  let testStorage: ReturnType<typeof createTestStorage>
-  let walletStorage: WalletStorageService
-
-  beforeEach(() => {
-    // REQUIRED: Create isolated test storage
-    testStorage = createTestStorage('my-test')
-    walletStorage = new WalletStorageService({ cwd: testStorage.configDir })
-  })
-
-  afterEach(() => {
-    // REQUIRED: Cleanup test directories
-    testStorage.cleanup()
-  })
-})
-```
-
-### Dangerous Pattern (FORBIDDEN):
-
-```typescript
-// ❌ NEVER DO THIS IN TESTS - touches real user config!
-const walletStorage = new WalletStorageService()
-walletStorage.getAllWallets().forEach(w => walletStorage.removeWallet(w.id)) // DELETES REAL DATA!
-```
-
-## Pre-Commit Checklist
-
-Run the following commands before committing:
-
-```bash
-npm run lint          # Check code style and potential issues
-npm run format        # Format code with Prettier
-npm run typecheck     # Run TypeScript type checking
-npm run test          # Run unit and integration tests
-```
-
-If any errors pop up, fix them before committing.
-
-## Development Workflow
-
-### Testing
-
-#### Unit Tests
-Unit tests are located in `src/tests/unit/` and cover:
-- Services (`src/services/*`)
-- Utilities (`src/utils/*`)
-- Storage (`src/storage/*`)
-
-Run unit tests:
-```bash
-npm test                    # Run all tests (excluding integration/e2e)
-npm test -- --watch         # Run tests in watch mode
-npm test -- --ui            # Run tests with Vitest UI
-```
-
-#### Integration Tests
-Integration tests are in `src/tests/integration/` and test:
-- Full workflows (wallet import, Safe creation, transaction lifecycle)
-- Service integration
-- Storage integration
-- Transaction building and parsing
-
-Run integration tests explicitly (they require blockchain access):
-```bash
-npm test src/tests/integration/integration-*.test.ts
-```
-
-#### E2E Tests
-E2E tests verify the CLI commands work correctly:
-- `e2e-cli.test.ts` - Basic CLI functionality
-- `e2e-wallet-commands.test.ts` - Wallet operations
-- `e2e-config-commands.test.ts` - Configuration management
-- `e2e-account-commands.test.ts` - Account operations
-- `e2e-tx-commands.test.ts` - Transaction commands
-- `integration-full-workflow.test.ts` - Complete end-to-end workflow
-
-Run E2E tests:
-```bash
-# Build the CLI first
-npm run build
-
-# Run E2E tests (requires TEST_WALLET_PK environment variable)
-TEST_WALLET_PK=0x... npm test src/tests/integration/e2e-*.test.ts
-```
-
-#### Coverage
-Check test coverage:
-```bash
-npm test -- --coverage       # Generate coverage report
-```
-
-Coverage thresholds are configured in `vitest.config.ts`:
-- Lines: 30%
-- Functions: 69%
-- Branches: 85%
-- Statements: 30%
-
-### Project Structure
-
-```
-src/
-├── commands/           # CLI command implementations (0% coverage - tested via E2E)
-│   ├── account/       # Safe account management
-│   ├── config/        # Configuration management
-│   ├── tx/            # Transaction operations
-│   └── wallet/        # Wallet management
-├── services/          # Business logic (87% coverage)
-│   ├── abi-service.ts
-│   ├── api-service.ts
-│   ├── contract-service.ts
-│   ├── ledger-service.ts
-│   ├── safe-service.ts
-│   ├── transaction-builder.ts
-│   ├── transaction-service.ts
-│   └── validation-service.ts
-├── storage/           # Data persistence (81% coverage)
-│   ├── config-store.ts
-│   ├── safe-store.ts
-│   ├── transaction-store.ts
-│   └── wallet-store.ts
-├── ui/                # CLI interface (0% coverage - interactive components)
-│   ├── components/
-│   ├── hooks/
-│   └── screens/
-├── utils/             # Utilities (96% coverage)
-│   ├── balance.ts
-│   ├── eip3770.ts
-│   ├── errors.ts
-│   ├── ethereum.ts
-│   └── validation.ts
-└── tests/
-    ├── fixtures/      # Test data and mocks
-    ├── helpers/       # Test utilities
-    ├── integration/   # Integration and E2E tests
-    └── unit/          # Unit tests
-```
-
-### Configuration and Storage
-
-If in the course of development or testing you need to clear or modify the local configs, back up the existing ones first, and restore them when finished.
-
-Configuration is stored in:
-- Config: `~/.config/@safe-global/safe-cli/config.json`
-- Data: `~/.local/share/@safe-global/safe-cli/`
-
-For testing with isolated directories, use `XDG_CONFIG_HOME` and `XDG_DATA_HOME`:
-```bash
-XDG_CONFIG_HOME=/tmp/test-config XDG_DATA_HOME=/tmp/test-data npm run dev
-```
-
-### Adding New Features
-
-1. **Create the service/utility** - Write the core logic with tests
-2. **Add storage layer** (if needed) - Implement data persistence
-3. **Create command** - Implement the CLI command in `src/commands/`
-4. **Add E2E test** - Verify the command works end-to-end
-5. **Update documentation** - Add to README if user-facing
-
-### Debugging
-
-Run CLI in development mode:
-```bash
-npm run dev -- <command>            # Run with tsx (fast reload)
-DEBUG=* npm run dev -- <command>    # Run with debug logging
-```
-
-Build and run production version:
-```bash
-npm run build
-node dist/index.js <command>
-```
-
-### Code Style
-
-- TypeScript strict mode enabled
-- ESLint for linting
-- Prettier for formatting
-- Husky for pre-commit hooks
-- lint-staged for staged file checking
-
-### Common Patterns
-
-#### Error Handling
-Use custom error classes from `src/utils/errors.ts`:
-```typescript
-import { ValidationError, SafeError } from '../utils/errors.js'
-
-throw new ValidationError('Invalid address format')
-throw new SafeError('Failed to create Safe')
-```
-
-#### Address Validation
-Support both plain and EIP-3770 addresses:
-```typescript
-import { parseEIP3770Address } from '../utils/eip3770.js'
-import { validateAddress } from '../utils/validation.js'
-
-const { chainId, address } = parseEIP3770Address('sep:0x...')
-validateAddress(address) // throws if invalid
-```
-
-#### Storage
-All storage services follow the same pattern:
-```typescript
-import { ConfigStore } from '../storage/config-store.js'
-
-const store = new ConfigStore()
-store.set('key', value)
-const value = store.get('key')
-```
-
-### Testing Best Practices
-
-1. **Isolate test data** - Use temporary directories for test configs/data
-2. **Mock external dependencies** - Mock API calls and blockchain interactions
-3. **Test error cases** - Verify error handling and edge cases
-4. **Use factories** - Use test helpers from `src/tests/helpers/factories.ts`
-5. **Clean up after tests** - Remove temporary files/directories in `afterEach`
-
-### Environment Variables
-
-- `TEST_WALLET_PK` - Private key for E2E tests (Sepolia testnet)
-- `XDG_CONFIG_HOME` - Custom config directory
-- `XDG_DATA_HOME` - Custom data directory
-- `NODE_ENV` - Set to 'test' during testing
-- `CI` - Set to 'true' for non-interactive mode
-
-### Blockchain Testing
-
-E2E tests that interact with blockchain require:
-- A funded Sepolia test wallet
-- `TEST_WALLET_PK` environment variable set
-- Network access to Sepolia RPC and Safe API
-
-Get Sepolia ETH:
-- [Sepolia Faucet](https://sepoliafaucet.com/)
-- [Alchemy Sepolia Faucet](https://sepoliafaucet.com/)
-
-### Troubleshooting
-
-**Tests timing out:**
-- Increase timeout in test: `{ timeout: 60000 }`
-- Check network connectivity
-- Verify RPC endpoints are accessible
-
-**Interactive prompts in tests:**
-- Use `CLITestHelper.execWithInput()` for tests with prompts
-- Set `CI=true` environment variable for non-interactive mode
-- Consider adding `--yes` flags to commands
-
-**Storage conflicts:**
-- Use isolated directories with `XDG_CONFIG_HOME` and `XDG_DATA_HOME`
-- Clean up in `afterEach` hooks
-- Use `mkdtempSync()` for temporary directories
+Read @CLAUDE.md