Skip to content

magic-dev-kz/decode-my-boss

Folders and files

NameName
Last commit message
Last commit date

Latest commit

Β 

History

24 Commits
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 

Repository files navigation

πŸ” Decode My Boss

What they say vs what they mean.

Paste a message from your boss β€” get the real meaning, red flags, toxicity score, and an actionable playbook. Plus: AI drafts your response. Free, no account needed.

Try it free β†’


Features

  • Decoded Translation β€” Every phrase analyzed: what was said vs what was meant, with tone tags (passive-aggressive, power move, genuine, etc.)
  • Red Flags β€” Detects patterns: passive aggression, gaslighting, blame shifting, weaponized positivity, and more
  • Toxicity Score (1-10) β€” Gradient bar with 5-category breakdown
  • Playbook β€” 2-3 specific, actionable steps for what to do next
  • ✍️ Draft Response β€” AI writes your reply, with tone control from Diplomatic to Direct
  • Quick Examples β€” click "Per my last email" or "Let's take this offline" to try instantly
  • Share Card β€” PNG card with score + flags (original text NOT included β€” privacy first)
  • Feedback Widget β€” tell us if the decode was helpful
  • BYOK (Bring Your Own Key) β€” works with OpenAI, Google Gemini, or Anthropic
  • Single HTML file β€” no backend, no account, no tracking

How It Works

  1. Paste a message from your boss / HR / colleague
  2. Select context (Corporate Email, Slack, Teams, HR Letter, Performance Review)
  3. Click "Decode"
  4. Get the translation, red flags, toxicity score, and playbook

Privacy

  • Your messages never touch our servers β€” text goes directly from your browser to the AI provider
  • API key stays in localStorage β€” we never see it
  • Share card contains NO original text β€” only score, flags, and verdict
  • No analytics, no cookies, no tracking

Supported AI Providers

Provider Model Cost per decode Notes
OpenAI gpt-4o-mini ~$0.01-0.03 Best quality
Google Gemini 2.0 Flash Free (free tier) Great for testing
Anthropic Claude Sonnet ~$0.01-0.03 CORS limitations in browser

Red Flag Types Detected

  • Passive Aggression
  • Gaslighting
  • Blame Shifting
  • Power Move
  • Exclusion Signal
  • False Urgency
  • Weaponized Positivity
  • Plausible Deniability
  • Scope Creep Trap
  • Credit Theft

Running Locally

# Simple HTTP server (needed for PWA)
python3 -m http.server 8000
# Visit http://localhost:8000

License

MIT β€” free for personal and commercial use.


Made by OpenClaw. We make $0 from this.

About

πŸ”“ AI decoder for corporate double-speak. Paste your boss's message, get the real meaning. Free, BYOK.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors