# Testing Jamf MCP Skills
## Overview
The Jamf MCP Server includes a comprehensive test suite for all skills that runs automatically before building. This ensures that all skills are working correctly before deployment.
## Running Tests
### Basic Test Run
```bash
npm test
# or
npm run test:skills
```
### Verbose Mode
To see detailed output including mock API calls and results:
```bash
VERBOSE=true npm run test:skills
```
### Build with Automatic Testing
The standard build command now includes automatic skill testing:
```bash
npm run build
```
### Build Without Testing
If you need to build quickly without running tests:
```bash
npm run build:force
```
## Test Coverage
The test suite covers all skills with the following test scenarios:
### 1. Device Search
- **Basic search**: Tests searching by device name
- **Possessive form**: Tests natural language like "Dwight's MacBook"
- **User search**: Tests searching by assigned user
### 2. Find Outdated Devices
- Tests identifying devices that haven't checked in within a specified timeframe
### 3. Batch Inventory Update
- Tests updating multiple devices concurrently
- Validates both successful and failed updates
### 4. Deploy Policy by Criteria
- Tests policy deployment with device criteria filtering
- Includes dry-run validation
### 5. Scheduled Compliance Check
- Tests comprehensive compliance auditing
- Validates summary report generation
### 6. Error Handling
- Tests graceful handling of invalid skill names
## Writing New Skill Tests
When adding a new skill, add corresponding tests to `test-all-skills.js`:
```javascript
await runner.runTest(
'Your Skill Name - Test Description',
'skill-name',
{
// Input parameters for the skill
param1: 'value1',
param2: 'value2'
},
[
// Assertions to validate the result
Object.assign(
(result) => result.success === true,
{ message: 'Should succeed' }
),
Object.assign(
(result) => result.data.someField === expectedValue,
{ message: 'Should return expected data' }
)
]
);
```
## CI/CD Integration
Tests run automatically in GitHub Actions on:
- Every push to main, develop, or feature branches
- Every pull request to main or develop
The CI pipeline tests against multiple Node.js versions (18.x, 20.x, 22.x) to ensure compatibility.
## Troubleshooting
### Tests Failing Locally
1. Ensure you have the latest dependencies: `npm install`
2. Rebuild the project: `npm run build:quick`
3. Run tests in verbose mode to see details: `VERBOSE=true npm test`
### Mock Data
The test suite uses mock Jamf API responses defined in `test-all-skills.js`. If you need to test with different data scenarios, modify the `mockDevices`, `mockPolicies`, or other mock data structures.
### Adding New Tools
If your skill uses a new tool, add it to the `createMockContext` function:
```javascript
case 'yourNewTool':
return {
data: {
// Mock response for your tool
}
};
```
## Best Practices
1. **Always run tests before committing**: The build process enforces this
2. **Keep tests fast**: Use mock data instead of real API calls
3. **Test edge cases**: Include tests for error conditions and empty results
4. **Update tests when modifying skills**: Keep tests in sync with skill behavior
5. **Use descriptive test names**: Make it clear what each test validates
