# Testing MCP Resources and Prompts
This document provides instructions for testing the MCP (Model Context Protocol) resources and prompts functionality in the Dataproc server.
## Overview
The Dataproc server now supports MCP resources and prompts, which allow LLM agents to:
1. **List available resources** - Get a list of tracked clusters and jobs
2. **Read resource details** - Get detailed information about a specific cluster or job
3. **List available prompts** - Get a list of available guidance prompts
4. **Read prompt content** - Get detailed guidance on specific topics
## Available Resources
The server exposes the following resources:
- **Cluster resources**: `dataproc://clusters/{projectId}/{region}/{clusterName}`
- **Job resources**: `dataproc://jobs/{projectId}/{region}/{jobId}`
## Available Prompts
The server provides the following prompts:
- **Dataproc Cluster Creation** (`dataproc-cluster-creation`): Guidelines for creating Dataproc clusters
- **Dataproc Job Submission** (`dataproc-job-submission`): Guidelines for submitting jobs to Dataproc clusters
## Running Tests
You can run the tests using the test runner script:
```bash
# Run all tests
node tests/run-tests.js all
# Run specific test types
node tests/run-tests.js unit
node tests/run-tests.js integration
node tests/run-tests.js simple
node tests/run-tests.js manual
```
## Test Types
### 1. Unit Tests
These tests verify the resource and prompt handler logic using Mocha and Chai. They use a CommonJS format (`.cjs` extension) to avoid module cycle issues.
```bash
npx mocha tests/unit/resource-handlers.test.cjs
```
### 2. Integration Tests
This script sets up a test MCP server with a custom memory transport and sends requests to test the resource and prompt handlers.
```bash
npx ts-node tests/integration/test-mcp-resources.ts
```
### 3. Simple Demo
This script demonstrates how the MCP resources and prompts handlers are set up and what they return. It doesn't require any external dependencies or actual clusters/jobs.
```bash
npx ts-node tests/manual/test-mcp-resources-simple.ts
```
### 4. Manual Tests
This script tests the resource and prompt handlers by sending requests to a running server. Start the server first, then run this script.
```bash
# Start the server in one terminal
node build/index.js
# Run the test script in another terminal
node tests/manual/test-mcp-resources.js
```
## Troubleshooting Common Issues
### Module Type Issues
This project uses ES modules (`"type": "module"` in package.json). Keep these points in mind:
1. **JavaScript files (.js)** are treated as ES modules and must use `import`/`export` syntax
2. **CommonJS files (.cjs)** can use `require()`/`module.exports` syntax
3. **TypeScript files (.ts)** follow the configuration in `tsconfig.json`
If you encounter module-related errors:
- Check that `.js` files use ES module syntax
- Use `.cjs` extension for files that need CommonJS syntax
- Make sure imports include the file extension (`.js`, `.cjs`, etc.)
### TypeScript Configuration
Make sure your `tsconfig.json` is properly configured for ES modules:
```json
{
"compilerOptions": {
"module": "NodeNext",
"moduleResolution": "NodeNext",
// other options...
}
}
```
### Running Tests with Mocha
When running tests with Mocha, use the appropriate command based on the file type:
- For CommonJS tests: `npx mocha tests/unit/resource-handlers.test.cjs`
- For TypeScript tests: `npx mocha --require ts-node/register tests/integration/test-mcp-resources.ts`
## Expected Responses
### List Resources Response
```json
{
"resources": [
{
"uri": "dataproc://clusters/project-id/region/cluster-name",
"name": "Cluster: cluster-name",
"description": "Dataproc cluster in region (RUNNING)"
},
{
"uri": "dataproc://jobs/project-id/region/job-id",
"name": "Job: job-id",
"description": "Dataproc job (submit_hive_query) - DONE"
}
]
}
```
### Read Resource Response (Cluster)
```json
{
"content": [
{
"type": "text",
"text": "{\"clusterName\":\"cluster-name\",\"status\":{\"state\":\"RUNNING\"},\"config\":{...}}"
}
]
}
```
### Read Resource Response (Job)
```json
{
"content": [
{
"type": "text",
"text": "{\"status\":{\"state\":\"DONE\"},\"results\":{...}}"
}
]
}
```
### List Prompts Response
```json
{
"prompts": [
{
"id": "dataproc-cluster-creation",
"name": "Dataproc Cluster Creation",
"description": "Guidelines for creating Dataproc clusters"
},
{
"id": "dataproc-job-submission",
"name": "Dataproc Job Submission",
"description": "Guidelines for submitting jobs to Dataproc clusters"
}
]
}
```
### Read Prompt Response
```json
{
"content": [
{
"type": "text",
"text": "# Dataproc Cluster Creation Guidelines\n\nWhen creating a Dataproc cluster, consider the following:\n\n1. **Region Selection**: Choose a region close to your data and users to minimize latency.\n..."
}
]
}
