modbender Humanize CLI
Analyze text for AI patterns and get actionable fixes. Helps content pass AI detectors by identifying robotic patterns.
Installation
npm install -g humanize-cli
Commands
Score Detection Risk
humanize score "Your text here"
humanize score -f article.txt
Returns 0-100% risk score:
- 0-20%: Low risk (appears human)
- 21-40%: Moderate risk
- 41-70%: High risk (likely flagged)
- 71-100%: Very high risk
Analyze AI Markers
humanize analyze "This comprehensive solution leverages cutting-edge technology."
Detects:
- AI vocabulary (delve, leverage, comprehensive, robust, seamless)
- Structural patterns (uniform sentences, perfect intro/conclusion)
- Missing human elements (no contractions, no opinions)
Get Improvement Suggestions
humanize suggest "It is important to note that our solution utilizes..."
Returns prioritized fixes with before/after examples.
Auto-Transform Text
humanize transform "The solution utilizes robust methodologies."
# Output: "The solution uses strong methods."
Automatically replaces AI vocabulary and adds human elements.
Watch Directory
humanize watch ./content --threshold 60
Monitor files and alert when detection risk exceeds threshold.
Common Use Cases
Check blog post before publishing:
humanize score -f blog-post.md
Fix AI-sounding content:
humanize transform -f draft.txt > improved.txt
Pipe from clipboard (macOS):
pbpaste | humanize suggest
Options
| Flag | Description |
|---|---|
-f, --file |
Read from file |
-j, --json |
JSON output |
-q, --quiet |
Minimal output |
-t, --threshold |
Alert threshold (0-100) |
What It Catches
AI Vocabulary:
- delve, leverage, utilize, comprehensive, robust, seamless
- stakeholder, synergy, actionable, paradigm
- furthermore, moreover, consequently, nevertheless
Structural Issues:
- No contractions (sounds robotic)
- Same-length sentences
- Perfect intro/body/conclusion
- Overuse of transitions
Built by LXGIC Studios