Setup & Installation
openclaw skills install @raghulpasupathi/gpt-analyzer
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
npx clawhub install gpt-analyzer
What This Skill Does
Analyzes text for GPT-generated content using phrase matching, structural patterns, and sentence uniformity checks. Attempts to distinguish between GPT-4 and GPT-3.5 outputs based on model-specific phrase signatures. Returns a confidence score, detected model, and a plain-language recommendation.
Unlike generic AI detectors, it attempts model-level attribution, separating GPT-4 patterns from GPT-3.5 ones rather than returning a single binary result.
When to use it
- Screening job applicant cover letters for AI authorship
- Checking student essay submissions for GPT usage
- Auditing blog or publication content before publishing
- Detecting AI-generated product reviews or support tickets
- Verifying whether a document draft was written by a human
Example Workflow
Here's how your AI assistant might use this skill in practice.
INPUT
User asks: Is this essay GPT-generated?
AGENT
- 1Receives the essay text as input
- 2Scans for GPT-4 and GPT-3.5 phrase signatures in the normalized text
- 3Analyzes structural patterns like numbered lists and bullet points
- 4Calculates sentence length variance to detect uniform writing style
- 5Returns confidence score, detected model version, and recommendation
OUTPUT
{ isGPT: true, confidence: 82, detectedModel: 'gpt-4', recommendation: 'Likely GPT' }