Documentation

Documentation

Application Methodology

Narrative Pro leverages advanced AI technology to transform how professionals create and manage their narratives. Our application is built on several key principles and features that work together to provide an intelligent, context-aware writing experience.

Core Components

Context-Aware Processing

Our AI analyzes your input within your specific professional context, whether military or civilian, ensuring relevant and accurate suggestions.

Intelligent Feedback

Comprehensive grading and feedback mechanisms help strengthen your narratives through detailed analysis and suggestions.

Secure Storage

The Narrative Vault leverages SSL encryption and Supabase's enterprise-grade security to safely store and manage your narratives.

AI-Powered Analysis

Our system uses advanced language models to provide contextual analysis based on military writing standards and best practices.

Modern Authentication

Our authentication system prioritizes both security and user experience through Magic Link and Google Auth:

Security Benefits
  • No password storage or management required
  • Eliminates risks of weak or reused passwords
  • Prevents credential stuffing attacks
  • Two-factor authentication built into the process
User Experience
  • No passwords to remember or reset
  • Seamless login with existing Google account
  • Simple one-click email authentication
  • Reduced friction during registration

Understanding Claude Tokens & API Keys

What are Claude Tokens?

Claude tokens are the fundamental units of text processing in the Anthropic Claude AI model. Think of them as the building blocks that make up your text - each token represents approximately 4 characters in English, though this can vary based on the specific content.

Token Impact on Processing

  • Processing Capacity: Determines the amount of text that can be analyzed in a single request
  • Response Length: Influences how detailed the AI's responses can be
  • Cost Calculation: Forms the basis for usage billing
  • Performance: Affects processing speed and efficiency

The Role of API Keys

API keys are secure credentials that enable safe communication between Narrative Pro and the Claude AI model. They serve multiple critical functions:

Security Features

  • Encrypted Communication
  • Access Control
  • Request Authentication
  • Usage Tracking

Management Benefits

  • Usage Monitoring
  • Resource Allocation
  • Billing Management
  • Access Revocation

Our Commitment to Transparency

We use Claude API keys directly to ensure complete token transparency. Unlike services that may skim tokens, with Narrative Pro:

  • You get 100% of what you pay for - every token you purchase goes directly to processing your narratives
  • Direct API key usage means transparent token management
  • No hidden token fees or deductions
  • Advanced features are provided through separate subscriptions, not token skimming

Token Management

Optimizing Token Usage

Best Practices

  • Keep narratives focused and concise
  • Use built-in templates
  • Monitor usage patterns
  • Regular usage reviews

Usage Monitoring

  • Track daily consumption
  • Set usage alerts
  • Review efficiency metrics
  • Analyze usage patterns

API Key Security

Protecting your API key is crucial for security and optimal performance:

  • Never share your API key with others
  • Rotate keys periodically (recommended every 90 days)
  • Monitor for unusual activity
  • Store keys securely in your account settings
  • Use environment variables for development

Cost Calculator

Claude API Pricing

Input Costs

  • Standard Input: $0.80/MTok
  • Prompt Cache Write: $1.00/MTok
  • Prompt Cache Read: $0.08/MTok

Output Costs

  • Generated Output: $4.00/MTok
  • 200K context window
  • 50% discount with Batches API

Typical Usage Cost Breakdown

Per Operation Cost

Based on typical usage patterns:

  • Narrative Generation Input: ~300 characters (≈ 75 tokens)
  • Grading Input: ~300 characters (≈ 75 tokens)
  • Output: 300-500 characters (≈ 75-125 tokens)

Cost Estimates

Single Operation:

  • Input Cost: ~$0.00012 (150 tokens × $0.80/MTok)
  • Output Cost: ~$0.0005 (125 tokens × $4.00/MTok)
  • Total per Operation: ~$0.00062

API Credit Examples

$10 API Credit:

≈ 8,064 operations

$50 API Credit:

≈ 40,322 operations

$100 API Credit:

≈ 80,645 operations

$500 API Credit:

≈ 403,225 operations

Getting Started

Basic Access - $5 API Credit:

Start with basic narrative generation:

  • Generate ≈ 4,032 narratives
  • Access to AI-powered narrative creation
  • Basic narrative storage
  • Cost per operation: ~$0.00062

Perfect for evaluating the core narrative generation capabilities.

Premium Features - Subscription Required:

Unlock advanced capabilities:

  • Narrative grading and feedback
  • Advanced Vault features
  • Enhanced analytics
  • Priority support

Subscription plans provide access to our full suite of professional tools, separate from API usage costs.

Frequently Asked Questions

What is the token limit per request?

The standard token limit is 100,000 tokens per request. This includes both your input text and the AI's response. For most narratives, this provides ample room for detailed analysis and suggestions.

How are tokens calculated?

Tokens are calculated based on the character count and complexity of your text. As a rough estimate, one token equals about 4 characters in English, but this can vary based on the specific content (numbers, special characters, etc.).

What happens if I exceed my token limit?

If you reach your token limit, the system will notify you and pause processing. You can either upgrade your plan for more tokens or wait for your allocation to reset (typically at the start of your billing cycle).

Can I transfer tokens between accounts?

Currently, tokens cannot be transferred between accounts. Each account's token allocation is specific to that account and resets according to the account's billing cycle.

How do I monitor my token usage?

You can monitor your token usage in real-time through your account dashboard. The dashboard provides detailed analytics including daily usage, historical trends, and remaining allocation.