Synapse - Dynamic Load Testing Tool

Synapse is a powerful command-line tool that generates K6 load testing scripts from simple YAML configurations. It supports dynamic URL construction, multiple parameter types including arrays, CSV data sources, and all K6 features.

🚀 Features

• Dynamic URL Construction - Build URLs with configurable parameters
• Multiple Parameter Types - Integer, string, array, CSV, and URL parameters
• Batch Mode - Use pre-built URLs from CSV files
• K6 Integration - Full K6 feature support with automatic script generation
• Smart Configuration - Automatic mode detection based on config
• Performance Metrics - Comprehensive load testing results
• CLI Interface - Easy-to-use command-line interface
• MCP Integration - Natural language load testing through LLM conversations
• 🆕 Image/Text Comparison - Compare responses between different versions/environments
• 🆕 Standalone Comparison - Direct CSV-based image/text comparison tool

📦 Installation

Global Installation (Recommended)

npm install -g synapse

Local Installation

npm install synapse
npx synapse --help

🏃 Quick Start

Simple Load Test (No Configuration File)

For quick testing without creating a YAML file:

synapse test --url "https://api.example.com" --concurrent 10 --requests 100

Advanced Configuration

1. Initialize Configuration

synapse init --name "My API Test" --url "https://api.example.com"

This creates a synapse.yml file with basic configuration.

2. Customize Configuration

Edit the generated synapse.yml:

name: "API Load Test"
baseUrl: "https://api.example.com/search"
execution:
  mode: "construct"
  concurrent: 10
  iterations: 100
parameters:
  - name: "query"
    type: "array"
    values: ["javascript", "python", "react & vue"]
  - name: "userId"
    type: "integer"
    min: 1000
    max: 9999
  - name: "imageUrl"
    type: "static"
    value: "http://localhost:7900/generated-content/imagen-3_0-generate-002/2025-11-05/imagen.webp"
k6Options:
  thresholds:
    http_req_duration: ["p(95)<500"]

Auto-encoding in action:

• "react & vue" → "react%20%26%20vue"
• Complex image URL → Automatically URL-encoded
• userId numbers → No encoding needed

3. Run Load Test

synapse run

📋 Parameter Types

Auto-Encoding: Static and CSV parameters automatically detect and URL-encode special characters and URLs. No manual encoding configuration needed for standard URL encoding.

Static Parameters ⭐ NEW

Use fixed values with smart auto-encoding for URLs and special characters:

- name: "imageUrl"
  type: "static"
  value: "http://localhost:7900/generated-content/imagen-3_0-generate-002/2025-11-05/imagen-1762381001933-3b19cf6c.webp"
- name: "apiKey"
  type: "static"
  value: "your-api-key-here"
- name: "searchQuery"
  type: "static"
  value: "hello world & special chars!"

Auto-encoding behavior:

• URLs are automatically URL-encoded when needed
• Special characters (&, !, spaces, etc.) are encoded
• Simple strings (like API keys) remain unchanged
• No configuration required

Integer Parameters

Generate random integers within a specified range:

- name: "userId"
  type: "integer"
  min: 1
  max: 1000000
  length: 10  # pad with zeros

String Parameters

Generate random strings with various character sets:

- name: "sessionId"
  type: "string"
  length: 32
  charset: "alphanumeric"  # or "alpha", "numeric", "custom"
  customChars: "abcdef123456"  # only if charset is "custom"

Array Parameters ⭐ NEW

Select random values from a predefined array with auto-encoding:

- name: "category"
  type: "array"
  values: ["electronics", "books & media", "home & garden"]
  unique: true  # optional: ensure no duplicates
- name: "testUrls"
  type: "array"
  values: [
    "http://localhost:3000/api/test?param=value",
    "https://example.com/search?q=hello world"
  ]

CSV Parameters

Load values from CSV files with smart auto-encoding:

- name: "region"
  type: "csv"
  file: "./data/regions.csv"
  column: "name"

URL Parameters

Load URLs with automatic encoding and optional base64:

- name: "targetUrl"
  type: "url"
  file: "./data/urls.csv"
  column: "url"
  encoding: "base64"  # Optional: only for base64 encoding

Note: URL encoding is automatic. Only specify encoding: "base64" when you need base64 encoding specifically.

🔄 Auto-Encoding Examples

Static Parameters:

# Input
- name: "imageUrl"
  type: "static"
  value: "http://localhost:7900/generated-content/imagen-3_0-generate-002/2025-11-05/imagen.webp"

# Output: Automatically encoded
# http%3A//localhost%3A7900/generated-content/imagen-3_0-generate-002/2025-11-05/imagen.webp

Array Parameters:

# Input
- name: "searchTerm"
  type: "array"
  values: ["hello world", "cats & dogs", "simple-text"]

# Output: Smart encoding
# "hello world" → "hello%20world"
# "cats & dogs" → "cats%20%26%20dogs"  
# "simple-text" → "simple-text" (unchanged)

CSV Data:

# urls.csv
url
http://localhost:3000/api/test?param=value&other=data
https://example.com/search?q=hello world
simple-endpoint

# Configuration
- name: "endpoint"
  type: "csv"
  file: "./urls.csv"
  column: "url"

# Output: Auto-encoded when needed
# Complex URLs → Encoded
# Simple strings → Unchanged

🔄 Comparison Features

Load Test with Comparison

Compare different versions/environments during load testing:

name: "Version Comparison Test"
baseUrl: "https://v1.api.example.com"
execution:
  mode: "construct"
  concurrent: 10
  iterations: 100
parameters:
  - name: "userId"
    type: "integer"
    min: 1000
    max: 9999
comparison:
  enabled: true
  type: "image"  # or "text"
  baseUrl2: "https://v2.api.example.com"
  threshold: 0.1
  timeout: 30000

Usage:

# Load test only (default)
synapse run

# Load test + comparison
synapse run --compare

# Comparison only (no load test)
synapse run --compare-only

Standalone Comparison

Compare images/text directly from CSV:

url1,url2
https://v1.api.com/image/1,https://v2.api.com/image/1
https://v1.api.com/image/2,https://v2.api.com/image/2

synapse compare --file comparison.csv --type image

Comparison Reports

Generated reports include:

• Response times for both URLs
• HTTP status codes
• Image dimensions and similarity scores
• Text match results
• Error details and timestamps
• CSV and JSON formats

🔧 CLI Commands

synapse test

Run simple load test without configuration file:

synapse test --url "https://api.example.com" --concurrent 10 --requests 100

Options:

• -u, --url <url> - Target URL to test (required)
• -c, --concurrent <number> - Number of concurrent users (required)
• -r, --requests <number> - Total number of requests (required)
• -o, --output <path> - Output directory (default: ./output)
• --dry-run - Generate script without running
• --keep-script - Keep generated K6 script

synapse run

Run load test from configuration:

synapse run --config synapse.yml --output ./results --dry-run

Options:

• -c, --config <path> - Configuration file path (default: synapse.yml)
• -o, --output <path> - Output directory (default: ./output)
• --dry-run - Generate script without running
• --keep-script - Keep generated K6 script
• --compare - Run comparison if enabled in config (opt-in)
• --compare-only - Generate URLs and run comparison only (skip load test)

synapse validate

Validate configuration file:

synapse validate --config synapse.yml

synapse generate

Generate K6 script without running:

synapse generate --config synapse.yml --output test.js

synapse init

Initialize new configuration:

synapse init --name "My Test" --url "https://api.example.com"

synapse compare

Compare images or text from CSV file:

synapse compare --file urls.csv --type image --column1 url1 --column2 url2

Options:

• -f, --file <path> - CSV file with URLs to compare (required)
• -c1, --column1 <name> - First URL column name (default: url1)
• -c2, --column2 <name> - Second URL column name (default: url2)
• -t, --type <type> - Comparison type: image or text (default: image)
• -o, --output <path> - Output directory (default: ./output)
• --timeout <ms> - Request timeout in milliseconds (default: 30000)
• --threshold <value> - Image comparison threshold 0-1 (default: 0.1)

📊 Execution Modes

Construct Mode

Dynamically builds URLs using base URL and parameters:

execution:
  mode: "construct"
  concurrent: 10
  iterations: 100

Batch Mode

Uses pre-built URLs from CSV file:

execution:
  mode: "batch"
  concurrent: 10
  duration: "5m"
batch:
  file: "./data/urls.csv"
  column: "url"

🎯 K6 Integration & Options

Synapse supports all K6 configuration options through the k6Options section. These options control test execution, performance thresholds, and advanced scenarios.

Basic Execution Options

k6Options:
  # Simple load test
  vus: 10              # Virtual users (concurrent connections)
  duration: "30s"      # Test duration
  iterations: 100      # Total number of iterations

📊 Thresholds - Performance Requirements

Thresholds define pass/fail criteria for your load test:

k6Options:
  thresholds:
    # Response time thresholds
    http_req_duration: ["p(95)<500"]           # 95% of requests under 500ms
    http_req_duration: ["p(99)<1000"]          # 99% of requests under 1s
    http_req_duration: ["avg<200"]             # Average response time under 200ms
    http_req_duration: ["med<150"]             # Median response time under 150ms
    
    # Error rate thresholds
    http_req_failed: ["rate<0.1"]              # Less than 10% failure rate
    http_req_failed: ["rate<0.01"]             # Less than 1% failure rate
    
    # Request rate thresholds
    http_reqs: ["rate>100"]                    # More than 100 requests/second
    
    # Check thresholds (custom validations)
    checks: ["rate>0.95"]                      # 95% of checks must pass
    
    # Data transfer thresholds
    data_received: ["rate<10000"]              # Less than 10KB/s received
    data_sent: ["rate<5000"]                   # Less than 5KB/s sent

Threshold Operators:

• < - Less than
• <= - Less than or equal
• > - Greater than
• >= - Greater than or equal
• == - Equal to
• != - Not equal to

Statistical Functions:

• avg - Average value
• min - Minimum value
• max - Maximum value
• med - Median (50th percentile)
• p(90) - 90th percentile
• p(95) - 95th percentile
• p(99) - 99th percentile
• rate - Rate (for counters)
• count - Total count

🎭 Scenarios - Advanced Test Patterns

Per-VU Iterations

Each virtual user runs a specific number of iterations:

k6Options:
  scenarios:
    api_test:
      executor: "per-vu-iterations"
      vus: 10
      iterations: 20        # Each VU runs 20 iterations (200 total)
      maxDuration: "5m"     # Maximum test duration

Shared Iterations

Total iterations shared among all VUs:

k6Options:
  scenarios:
    load_test:
      executor: "shared-iterations"
      vus: 10
      iterations: 1000      # 1000 total iterations shared
      maxDuration: "10m"

Constant VUs

Fixed number of VUs for a duration:

k6Options:
  scenarios:
    steady_load:
      executor: "constant-vus"
      vus: 50
      duration: "5m"

Ramping VUs

Gradually increase/decrease load:

k6Options:
  scenarios:
    ramp_test:
      executor: "ramping-vus"
      startVUs: 0
      stages:
        - duration: "2m"
          target: 10        # Ramp up to 10 VUs
        - duration: "5m"
          target: 50        # Ramp up to 50 VUs
        - duration: "2m"
          target: 0         # Ramp down to 0 VUs

Constant Arrival Rate

Maintain steady request rate:

k6Options:
  scenarios:
    constant_rate:
      executor: "constant-arrival-rate"
      rate: 100            # 100 iterations/second
      duration: "5m"
      preAllocatedVUs: 10  # Pre-allocated VUs
      maxVUs: 50           # Maximum VUs if needed

Ramping Arrival Rate

Variable request rate over time:

k6Options:
  scenarios:
    variable_rate:
      executor: "ramping-arrival-rate"
      startRate: 10        # Start at 10 iter/sec
      stages:
        - duration: "2m"
          target: 50       # Ramp to 50 iter/sec
        - duration: "5m"
          target: 100      # Ramp to 100 iter/sec
        - duration: "2m"
          target: 0        # Ramp down to 0
      preAllocatedVUs: 20
      maxVUs: 100

🔧 Advanced Options

k6Options:
  # Global settings
  userAgent: "MyApp LoadTest/1.0"
  
  # HTTP settings
  http:
    timeout: "60s"
    keepAlive: true
  
  # TLS settings
  tlsAuth:
    - domains: ["example.com"]
      cert: "client.crt"
      key: "client.key"
  
  # DNS settings
  dns:
    ttl: "5m"
    select: "random"      # or "roundRobin", "first"
  
  # Rate limiting
  rps: 500               # Requests per second limit
  
  # Batch settings
  batch: 10              # Batch HTTP requests
  batchPerHost: 5        # Batch per host
  
  # Connection settings
  maxRedirects: 10
  userAgent: "k6/0.45.0"
  
  # Output settings
  summaryTrendStats: ["avg", "min", "med", "max", "p(90)", "p(95)", "p(99)"]

📈 Complete Example

name: "Production API Load Test"
baseUrl: "https://api.example.com"
execution:
  mode: "construct"
  concurrent: 50
  iterations: 5000

parameters:
  - name: "userId"
    type: "integer"
    min: 1000
    max: 9999

k6Options:
  # Performance requirements
  thresholds:
    http_req_duration: ["p(95)<500", "p(99)<1000"]
    http_req_failed: ["rate<0.01"]
    http_reqs: ["rate>200"]
    checks: ["rate>0.99"]
  
  # Load pattern
  scenarios:
    normal_load:
      executor: "ramping-vus"
      stages:
        - duration: "5m"
          target: 10      # Warm up
        - duration: "10m"
          target: 50      # Normal load
        - duration: "5m"
          target: 100     # Peak load
        - duration: "10m"
          target: 50      # Back to normal
        - duration: "5m"
          target: 0       # Cool down
  
  # Global settings
  userAgent: "LoadTest/1.0"
  maxRedirects: 5
  
  # HTTP configuration
  http:
    timeout: "30s"
    keepAlive: true

🚨 Common Threshold Patterns

API Performance:

thresholds:
  http_req_duration: ["p(95)<200", "p(99)<500"]
  http_req_failed: ["rate<0.01"]

High Throughput:

thresholds:
  http_reqs: ["rate>1000"]
  http_req_duration: ["p(95)<100"]

Reliability Focus:

thresholds:
  http_req_failed: ["rate<0.001"]  # 99.9% success rate
  checks: ["rate>0.999"]

📁 Project Structure

synapse/
├── src/
│   ├── generators/          # Parameter and script generators
│   ├── validators/          # Configuration validators
│   ├── types/              # TypeScript interfaces
│   └── cli.ts              # CLI interface
├── examples/               # Example configurations
├── docs/                   # Documentation
└── tests/                  # Test files

🧪 Testing

Run the test suite:

npm test

Run with coverage:

npm run test -- --coverage

📚 Documentation

Full documentation is available at: Synapse Docs

Or serve locally:

npm run docs:serve

🤝 Contributing

1. Fork the repository
2. Create a feature branch: git checkout -b feature/amazing-feature
3. Commit changes: git commit -m 'Add amazing feature'
4. Push to branch: git push origin feature/amazing-feature
5. Open a Pull Request

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

🙏 Acknowledgments

• K6 - Modern load testing tool
• Commander.js - CLI framework
• Joi - Schema validation

🐛 Issues & Support

• Report bugs: GitHub Issues
• Feature requests: GitHub Discussions
• Documentation: Synapse Docs