Setup

Getting Started

Everything you need to know about authenticating and integrating with the Scripthuman API.

1. Generating your API Key

To make requests, you need a secret key.

Log in to your Scripthuman Dashboard.
Navigate to Settings > API Keys.
Click "Generate New Secret Key".
Copy the key (sk_live_...). Store it safely; you won't see it again.

Warning: Never share your API key in client-side code (browsers/apps). Only use it on your server.

2. Authentication Headers

Include the key in the Authorization header of every request:

Authorization:Bearer sk_live_51Mz...

Rate Limits & Errors

Status	Code	Description
200 OK	Success	Request processed successfully.
401	Unauthorized	Invalid or missing API key.
429	Too Many Requests	You exceeded your plan's rate limit (Free: 60 req/min).

Core Tool

The Writing Room

The Writing Room is your command center for detecting "Silicon Smog" and restoring human essence to robotic text.

1AI Detection Heatmap

When you paste text, our engine scans for statistical anomalies typical of LLMs (ChatGPT, Claude, etc.). We visualize this using a traffic-light heatmap:

Red (High AI)

Monotone sentence structures. Very low burstiness. Likely 100% AI.

Yellow (Mixed)

Possible human edits, but underlying robotic syntax remains.

Green (Human)

Natural variance and complexity. Passes all major detectors.

2Humanization Modes

The "Humanize" button doesn't just rephrase; it reconstructs the cognitive pattern of the text based on your selected mode:

Standard (Default)

Balanced approach. Maintains the original message clarity while disrupting AI syntax patterns. Best for general web content.

Academic

Preserves formal terminology, citations, and structural rigor. Increases complexity without sacrificing professionalism. Ideal for essays and papers.

Creative / Storyteller

Maximizes metaphor usage and sentence variety. Introduces higher unpredictability (perplexity) to mimic human storytelling flair.

Advanced

Style Studio

Fine-tune the nuances of your writing using our Vibe Audit engine and Locale Injectors.

Vibe Audit Metrics

1. Perplexity

Measures the "unpredictability" of a text. AI models are trained to minimize perplexity (be predictable). Humans are naturally chaotic.

High Perplexity: Good. Indicates unique word choices.
Low Perplexity: Bad. Indicates generic, robotic phrasing.

2. Burstiness

Measures the variation in sentence structure and length. AI tends to write sentences of average length and similar structure (monotone).

Goal: Mix short, punchy sentences with longer, complex clauses.

Locale Injection

Don't just translate; localized. This tool injects region-specific idioms, slang, and cultural references into your text to make it feel native.

Supported Locales:

London Tech

NYC Finance

Aussie Casual

Valley VC

Legal Standard

Gen-Z Internet

Agency

Enterprise Hub

Configuration for large teams. Setup brand guardrails and maintain voice consistency at scale.

Configuring Brand Guardrails

Guardrails allow you to blacklist specific words, enforce preferred terminology, and set tone restrictions for all team members. These are defined via a JSON configuration in the admin dashboard.

// Example Guardrail Configuration
{
  "banned_words": ["synergy", "deep dive", "leverage"],
  "voice_settings": {
    "max_sentence_length": 25,
    "allow_passive_voice": false
  }
}

Validation

Trust Hub

Verify content legitimacy. Our 4-layer forensic engine provides proof of human authorship.

Verification Badge

Embed this live badge on your blog or article. It fetches real-time validation status from our servers.

The 4-Layer Analysis

1
Stylometric Analysis: Checks for statistical patterns of AI usage.
2
Entropy Check: Measures randomness and chaos in sentence construction.
3
Watermark Detection: Scans for known watermarks from OpenAI/Anthropic.
4
Fact Consistency: Verifies internal logical consistency of the narrative.