N.L.D.S. User Guide - Getting Started

Welcome to N.L.D.S.

The Natural Language Detection System (N.L.D.S.) is your gateway to the JAEGIS Enhanced Agent System. This guide will help you get started with natural language processing, command generation, and intelligent task automation.

What is N.L.D.S.?

N.L.D.S. is the Tier 0 component of JAEGIS that transforms natural language input into precise, executable commands. It understands your intent across multiple dimensions and generates optimized instructions for the JAEGIS agent system.

Key Capabilities

• Multi-dimensional Analysis: Logical, emotional, and creative understanding

• High Confidence Processing: ≥85% accuracy with alternative interpretations

• Real-time Processing: <500ms response time for most requests

• Intelligent Command Generation: Automatic JAEGIS command optimization

• Context Awareness: Session-based learning and adaptation

Quick Start

1. Authentication Setup

First, obtain your API credentials:

# Using API Key (Simple)
export NLDS_API_KEY="your-api-key-here"

# Using JWT Token (Recommended)
export NLDS_JWT_TOKEN="your-jwt-token-here"

2. Your First Request

Try your first N.L.D.S. request:

curl -X POST "https://api.jaegis.ai/v2/process" \
  -H "Authorization: Bearer $NLDS_JWT_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "input_text": "Create a secure user authentication system",
    "user_id": "getting-started-user"
  }'

Response:

{
  "request_id": "req_abc123",
  "overall_confidence": 0.91,
  "primary_command": {
    "command": "FRED:IMPLEMENT --component=auth --security=high",
    "confidence": 0.91,
    "squad": "development"
  }
}

3. Understanding the Response

• overall_confidence: How confident N.L.D.S. is in its understanding (0-1)

• primary_command: The main JAEGIS command generated

• squad: Which JAEGIS squad should handle the task

• alternative_interpretations: Other possible interpretations

Core Concepts

Processing Dimensions

N.L.D.S. analyzes your input across three dimensions:

Logical Dimension

• Technical requirements and constraints

• System architecture considerations

• Implementation complexity assessment

{
  "processing_dimensions": ["logical"],
  "input_text": "Build a REST API with authentication"
}

Emotional Dimension

• User sentiment and urgency

• Satisfaction factors and concerns

• Communication tone and style

{
  "processing_dimensions": ["emotional"],
  "input_text": "I urgently need help fixing this critical bug!"
}

Creative Dimension

• Alternative approaches and solutions

• Innovation opportunities

• Design patterns and best practices

{
  "processing_dimensions": ["creative"],
  "input_text": "Design an innovative user onboarding experience"
}

All Dimensions (Default)

{
  "processing_dimensions": ["all"],
  "input_text": "Create a user-friendly dashboard"
}

Confidence Levels

N.L.D.S. provides confidence scoring for all interpretations:

Confidence Range	Level	Action
85-100%	High	Direct execution recommended
70-84%	Medium	Review recommended
50-69%	Low	Clarification needed
0-49%	Very Low	Rephrase required

JAEGIS Integration

N.L.D.S. generates commands for different JAEGIS squads:

Squad	Purpose	Example Commands
development	Software development	FRED:IMPLEMENT, FRED:BUILD
analysis	Research and analysis	TYLER:ANALYZE, TYLER:INVESTIGATE
security	Security tasks	SECURE:PROTECT, SECURE:AUDIT
content	Documentation	DOCUMENT:CREATE, DOCUMENT:UPDATE
integration	System integration	INTEGRATE:CONNECT, INTEGRATE:SYNC

Common Use Cases

1. Software Development

Input: "Create a microservice for user management with Docker deployment"

N.L.D.S. Processing:

{
  "input_text": "Create a microservice for user management with Docker deployment",
  "processing_dimensions": ["logical", "creative"],
  "preferred_squad": "development"
}

Expected Output:

{
  "primary_command": {
    "command": "FRED:IMPLEMENT --service=user_management --architecture=microservice --deployment=docker",
    "mode": 3,
    "squad": "development"
  },
  "alternative_interpretations": [
    {
      "command": "ARCHITECT:DESIGN --service=user_management --pattern=microservice",
      "reasoning": "Design-first approach for complex microservice"
    }
  ]
}

2. System Analysis

Input: "Analyze database performance and identify slow queries"

N.L.D.S. Processing:

{
  "input_text": "Analyze database performance and identify slow queries",
  "processing_dimensions": ["logical"],
  "preferred_squad": "analysis"
}

Expected Output:

{
  "primary_command": {
    "command": "TYLER:ANALYZE --target=database --metrics=performance --focus=slow_queries",
    "mode": 2,
    "squad": "analysis"
  }
}

3. Security Assessment

Input: "Perform security audit on the authentication system"

N.L.D.S. Processing:

{
  "input_text": "Perform security audit on the authentication system",
  "processing_dimensions": ["logical"],
  "preferred_squad": "security"
}

Expected Output:

{
  "primary_command": {
    "command": "SECURE:AUDIT --component=authentication --scope=comprehensive",
    "mode": 4,
    "squad": "security"
  }
}

Best Practices

1. Writing Effective Prompts

Good Examples:

• ✅ "Create a REST API for user authentication with JWT tokens"

• ✅ "Analyze system performance metrics and identify bottlenecks"

• ✅ "Deploy the application to production with zero downtime"

Avoid:

• ❌ "Do something"

• ❌ "Fix it"

• ❌ "Make it better"

2. Providing Context

Include relevant context for better results:

{
  "input_text": "Optimize database queries",
  "context": {
    "database_type": "postgresql",
    "current_performance": "slow",
    "priority": "high",
    "environment": "production"
  }
}

3. Using Sessions

Maintain context across requests:

{
  "input_text": "Now add rate limiting to the API",
  "session_id": "session_123",
  "user_id": "user_456"
}

4. Handling Low Confidence

When confidence is low, try:

1. Add more details: "Create a secure user authentication system with JWT tokens and role-based access control"

2. Specify technology: "Build a Python Flask API with PostgreSQL database"

3. Include context: "For a web application with 1000+ users"

Advanced Features

Batch Processing

Process multiple requests efficiently:

{
  "inputs": [
    {
      "id": "task_1",
      "text": "Create user authentication system"
    },
    {
      "id": "task_2", 
      "text": "Set up monitoring and logging"
    },
    {
      "id": "task_3",
      "text": "Deploy to production environment"
    }
  ],
  "batch_options": {
    "parallel_processing": true
  }
}

Custom Preferences

Set user preferences for consistent results:

{
  "input_text": "Create a web application",
  "user_preferences": {
    "default_technology": "python",
    "architecture_preference": "microservices",
    "security_level": "high",
    "deployment_target": "cloud"
  }
}

Streaming Responses

For real-time processing updates:

const eventSource = new EventSource('/api/v2/stream/process');
eventSource.onmessage = function(event) {
  const data = JSON.parse(event.data);
  console.log('Processing update:', data);
};

Troubleshooting

Common Issues

Low Confidence Results

Problem: Confidence < 70% Solution:

• Add more specific details

• Include technical requirements

• Provide context information

Rate Limiting

Problem: 429 Too Many Requests Solution:

• Check your tier limits

• Implement exponential backoff

• Consider upgrading your plan

Timeout Errors

Problem: Request timeout Solution:

• Simplify complex requests

• Use batch processing for multiple tasks

• Increase timeout settings

Getting Help

1. Check Status: https://status.jaegis.ai

2. Documentation: https://docs.jaegis.ai

3. Support: [e-mail]

4. Community: GitHub Discussions

Next Steps

1. Explore SDKs: Try the Python SDK or JavaScript SDK

2. Advanced Tutorials: Learn about complex workflows

3. Integration Patterns: Discover integration best practices

4. API Reference: Review the complete API documentation

Examples Repository

Find more examples in our GitHub repository:

• Basic Examples

• Advanced Use Cases

• Integration Samples

---

Need Help? Contact our support team at [e-mail] or visit our community forum.