AI4U LLC · AI4UTECH.COM · ROUTING OS v3.0
AI4U Right-Tool
Routing OS
Route every task to the right AI. Prove the result.
Preserve the artifact. Learn from every win or failure.
The world's first AI operating system for consultants, builders, and operators who refuse to guess which AI to use.
See it in action
How it works
Tap any step above to see what it does.
Try a real example
14
Platforms Mapped
8
Task Stages
4
Risk Levels
∞
Routing Rules
100%
Free to Use
Your Task
→
Routing OS
→
Stage + Risk
→
Platform Match
→
Optimized Prompt
→
AI Output
→
Proof Receipt
→
Log & Learn
Live Intelligence
Toolchain Confidence Map
Platform Confidence Scores
Proof Maturity Score
Route Success by Task Stage
V1 → V6 Roadmap Progress
Operating Doctrine
The AI4U Self-Improvement Loop
Route
Decision
Decision
→
Platform
Payload
Payload
→
Proof
Receipt
Receipt
→
Outcome
Record
Record
→
Pattern
Extraction
Extraction
→
Failure
Immunization
Immunization
→
Better
Future Route
Future Route
Start Here
Not Sure Which AI To Use?
Start Here.
Describe what you want done. AI4U picks the best tool, writes the instructions, and shows you exactly what to do next.
🎯
Pick the right AI
Stop guessing between ChatGPT, Claude, Gemini, and more
✍️
Get the exact words to use
AI4U writes the prompt for you
🗺️
Know where to start
We show you what to click and what to type
📚
Save what worked
Build your own AI playbook over time
Try It Now
Pick a real task to see how it works
Click any card — AI4U fills in the Route Engine and shows you exactly which AI to use and what to say.
📧
Write a professional email
Writing · Low risk
📄
Improve my resume
Writing · Medium risk
🔍
Compare products before buying
Research · Low risk
🌐
Build a website or app
Building · Medium risk
🔧
Fix something that isn't working
Repair · High risk
⚙️
Automate a repetitive task
Automation · Medium risk
📢
Create content for my business
Content · Low risk
🧪
Research something deeply
Research · Medium risk
Step by Step
How AI4U Works — Plain English
Quick Reference
Task → Platform Routing Guide
Writing & Content
| Blog Posts / Articles | Claude Sonnet |
| Marketing Copy | ChatGPT-4o |
| Technical Docs | Claude Opus |
| Email Sequences | Claude Sonnet |
| Presentations | Manus Agent |
Code & Automation
| Full-Stack Apps | Manus Agent |
| Repo Repair / Hardening | Claude Code |
| PR / Code Tasks | Codex CLI |
| Durable Automation | Trigger.dev |
| No-Code Workflows | Zapier |
Research & Analysis
| Real-time Research | Perplexity Pro |
| Deep Analysis | Gemini Ultra |
| Data Visualization | Manus Agent |
| Competitive Intel | Perplexity |
| Google-Native Grounding | Vertex AI |
Infrastructure & Proof
| Public App Deployment | Vercel |
| Auth / DB / Vectors | Supabase |
| Source of Truth | GitHub |
| CI / Proof Receipts | GitHub Actions |
| AI Evals | promptfoo |
Plain English Glossary
What All These Terms Actually Mean
The tool that analyzes your task and recommends the single best AI platform for the job based on where you are and how much is at stake.
Where you are in your project — whether you're starting an idea, building a draft, fixing something broken, deploying to the public, or automating a process.
How much accuracy matters for this task — from freely exploring ideas all the way to mission-critical work where an error cannot happen.
The tool that writes the exact instructions for the AI platform AI4U recommended — ready to copy and paste, no prompt engineering required.
A checklist of evidence that confirms your AI task actually worked — things like a live URL loading, a build passing, or a document being complete and accurate.
Your personal log of every AI task you've run — what worked, what didn't, and why — so you can duplicate successes and avoid repeating failures.
A reference grid showing all 14 AI platforms AI4U tracks, what each one is best at, and when to use it — so you can compare options at a glance.
Routing Cockpit
AI4U Route Decision Engine
Define your task, select stage and risk level, and receive an intelligent platform recommendation with a full routing payload.
Task Parameters
Describe what you want done in normal language
What you already have (optional)
What's blocking you (optional)
What you want back
Quick Route by Category
Writing
Coding
Research
Deploy
Automate
Analyze
Design
Proof
Awaiting Route Parameters
Fill in the form and click Route This Task
Routing Intelligence
Platform Selection Rules
Idea → Prototype
Use Manus Agent for big-idea-to-first-prototype. It handles multi-step orchestration, file creation, and deployment in a single session.
Prototype → Runnable
Use Replit Agent for interactive, runnable prototypes with live preview. Best for demos and client-facing MVPs.
Broken Codebase
Use Claude Code for repo repair and hardening. It reads entire codebases and applies surgical fixes with full context.
PR / Code Tasks
Use Codex CLI as a PR worker. Best for isolated, well-defined code tasks with clear acceptance criteria.
Production Deploy
Use Vercel for public app deployment. Zero-config, instant CDN, custom domains, and GitHub CI/CD built in.
Mission Critical
Use Claude Opus + Supabase + GitHub Actions for mission-critical work. Every step must produce a verifiable proof receipt.
Intelligence Grid
AI4U Platform Matrix
Every platform in the AI4U ecosystem — mapped by use case, strength, and when to deploy it.
Manus Agent
Big Idea → First Prototype
Multi-step AI agent that orchestrates full projects — research, code, deploy, document — in a single session.
Use when: You have a vision and need a complete first build
Orchestration
Replit Agent
Runnable Prototype
Live-preview coding environment with AI agent. Best for interactive demos, MVPs, and client-facing prototypes.
Use when: You need a live, shareable demo fast
Prototype
Claude Code
Repo Repair / Hardening
Deep codebase reader and surgical fixer. Reads entire repos and applies precise, context-aware changes.
Use when: Existing code is broken, messy, or needs hardening
Repair
Codex CLI
PR / Code Task Worker
Isolated code task executor. Best for well-defined, scoped tasks with clear acceptance criteria.
Use when: You have a specific, bounded code task
Code Tasks
Vercel
Public App Deployment
Zero-config deployment platform with instant CDN, custom domains, and GitHub CI/CD. The gold standard for static and Next.js apps.
Use when: You need a live public URL with automatic deploys
Deploy
Supabase
Auth · Database · Memory · Vectors · Proof
Open-source Firebase alternative with Postgres, real-time subscriptions, auth, storage, and vector embeddings.
Use when: You need persistent data, user auth, or proof receipts
Backend
GitHub
Source of Truth
The canonical source of truth for all code. Every project must have a GitHub repo before it's considered real.
Use when: Any code exists — always
Version Control
GitHub Actions
CI / Proof Receipts
Automated workflows triggered by code pushes. Used for CI/CD, automated testing, and generating proof receipts.
Use when: You need automated verification on every deploy
Automation
promptfoo
AI Evals
Open-source LLM evaluation framework. Test prompts, compare models, and generate quantitative proof of AI output quality.
Use when: You need to prove AI output quality objectively
Evaluation
Sentry / LangSmith
Observability
Error tracking and LLM trace monitoring. Sentry for app errors; LangSmith for LLM chain observability and debugging.
Use when: You need to monitor production AI systems
Monitoring
Zapier
Simple No-Code Workflow
No-code automation platform. Best for simple, linear workflows connecting SaaS tools without custom code.
Use when: You need a quick automation with no coding
No-Code
Trigger.dev
Durable Automation
Code-first background job platform with durable execution, retries, and long-running task support.
Use when: You need reliable, code-based background automation
Automation
Wix
AI4Utech.com Funnel / Site
The platform powering AI4Utech.com. Handles the main consulting funnel, contact forms, and brand presence.
Use when: Managing the AI4U main website and client funnel
Website
Vertex AI
Google-Native Grounding
Google Cloud's AI platform with Gemini models, grounding with Google Search, and enterprise-grade safety controls.
Use when: You need Google-native AI with real-time grounding
Enterprise AI
Decision Matrix
Platform × Task Stage
| Platform | Idea | Proto | Repair | Deploy | Auto | Agent | Proof | Biz |
|---|
Prompt Intelligence
AI4U Prompt Generator
Generate expert-level, platform-specific prompts that extract maximum performance from any AI tool.
Prompt Parameters
Quick Templates
Full-Stack App
Manus Agent build prompt
Code Repair
Claude Code fix prompt
Deep Research
Perplexity research prompt
Content Creation
Claude writing prompt
Deploy & Verify
Vercel + GitHub prompt
AI Agent Build
Agent architecture prompt
Generated Prompt
Your generated prompt will appear here.
Fill in the parameters on the left and click Generate Prompt.
The output will include:
→ Router Summary
→ Toolchain Selected
→ Where to Paste This
→ Exact Prompt / Instructions
→ Verification Steps
→ Handoff / Escalation Rule
Verification Protocol
AI4U Proof Chain
Every AI output must pass through the Proof Chain before it's considered complete. No proof = no done.
This Site — Deployment Proof
Primary Production URL
https://ai4u-routing-os.vercel.app →
Custom Domain (Live ✓)
https://routing.ai4utech.com →
Vercel Dashboard
Vercel Project →
GitHub Source of Truth
leephanna/ai4u-routing-os →
Proof Checklist — v3.0
| # | Check | Status |
|---|---|---|
| 1 | routing.ai4utech.com loads | ✓ PASS |
| 2 | ai4u-routing-os.vercel.app loads | ✓ PASS |
| 3 | HTML renders without console errors | ✓ PASS |
| 4 | AI4U badge displays and spins | ✓ PASS |
| 5 | Desktop navigation works | ✓ PASS |
| 6 | Mobile hamburger menu works | ✓ PASS |
| 7 | Route engine changes output | ✓ PASS |
| 8 | Prompt generator creates prompt | ✓ PASS |
| 9 | Copy button works | ✓ PASS |
| 10 | Ledger add/save works (localStorage) | ✓ PASS |
| 11 | Slideshow controls work | ✓ PASS |
| 12 | Charts display | ✓ PASS |
| 13 | All links work | ✓ PASS |
| 14 | Image assets served correctly (not HTML) | ✓ PASS |
| 15 | Static deploy compatible | ✓ PASS |
| 16 | No secrets in code | ✓ PASS |
| 17 | GitHub repo clean | ✓ PASS |
Proof Persistence
Save a Proof Receipt
Record a proof artifact to your Supabase ledger. Works locally if not signed in.
Proof Protocol
The 7-Link Proof Chain
1
Browser Proof
The output loads in a real browser at a real URL. Screenshot the live page. Verify HTTP 200. Test in incognito mode. Record the URL and timestamp.
✓ Verified
2
Build Proof
The deployment pipeline completed without errors. Vercel shows "Ready" status. GitHub shows the commit SHA. Build logs are clean.
✓ Verified
3
Link Proof
All internal and external links resolve correctly. No 404s. Custom domain points to the correct deployment. CDN assets load.
✓ Verified
4
Data Persistence Proof
Data survives a page refresh. localStorage entries persist. (V2: Supabase records survive a session end and return on re-login.)
⏳ V2 — Supabase Upgrade
5
Email / Automation Proof
Automated workflows trigger correctly. Emails arrive. Webhooks fire. (V3: GitHub Actions proof receipts on every deploy.)
⏳ V3 — GitHub Actions
6
Screenshot / Log Proof
Visual evidence exists. Screenshots are timestamped. Console logs are clean. Error monitoring shows no critical alerts.
✓ Verified
7
Failure Fallback Proof
A documented fallback platform exists if the primary fails. The fallback has been tested. Escalation rules are defined and recorded.
✓ Documented
Permanent Learning System
AI Task Ledger
Log New Entry
Ledger Entries
📊
No entries yet
Log your first task outcome to start building your AI performance database.
Learning Intelligence
—
Outcomes
—
Win Rate %
—
Top Platform
—
Most Failed
Local data
Data Source
Evolution Path
AI4U Routing OS Roadmap
From static command center to full AI operating intelligence. Each version unlocks a new layer of capability.
Version Roadmap
V1
V1 — LIVE NOW
Static Public Command Center
Route engine, prompt generator, proof chain, localStorage ledger, platform matrix, slideshow, charts. Deployed on Vercel with custom domain.
✓ Deployed
V2
V2 — NEXT
Supabase Persistent Ledger
Replace localStorage with Supabase Postgres. Ledger entries survive browser clears. User auth. Cross-device sync. Team sharing.
⏳ Planned
V3
V3 — PLANNED
GitHub Actions Proof Automation
Automated proof receipts on every deploy. CI pipeline generates proof artifacts. Proof chain becomes fully automated and auditable.
⏳ Planned
V4
V4 — FUTURE
AI-Assisted Routing Recommendations
LLM-powered routing engine that learns from your ledger. Personalized recommendations based on your historical success rates.
🔮 Future
V5
V5 — FUTURE
Cross-Platform Outcome Intelligence
Aggregate routing outcomes across all AI4U clients. Community-powered routing intelligence. Pattern library shared across the ecosystem.
🔮 Future
V6
V6 — VISION
AI4U Client-Facing Version
White-labeled routing OS for AI4U consulting clients. Custom branding, private ledgers, team dashboards, and AI4U consulting integration.
🚀 Vision
Next 3 Actions
1
Add Supabase Persistent Ledger
Replace localStorage with Supabase Postgres for cross-device, permanent storage
2
Add GitHub Actions Proof Workflow
Automate proof receipts on every push to main — build, test, and verify automatically
3
Add AI-Powered Routing Recommendations
Connect to Gemini or Claude API to generate dynamic routing suggestions from your ledger history
Ready to Upgrade?
Lee Hanna and the AI4U team can implement V2 Supabase persistence, V3 GitHub Actions automation, and custom routing intelligence for your organization.
Contact AI4U LLC →Tech Stack
Static HTML
Vanilla JS
CSS3 Animations
Chart.js CDN
Vercel Deploy
GitHub Source
Orbitron Font
Inter Font
localStorage
Zero Dependencies
Configuration
OS Settings
Preferences
Default Platform
Pre-select in Route Engine and Prompt Gen
Default Risk Level
Pre-select in Route Engine
Your Name
Used in generated prompts
Organization
Used in generated prompts
About This OS
AI4U Routing OS
Version 3.0 — Static Edition
© 2026 AI4U, LLC
Built by Lee Hanna, AI Consultant and Owner of AI4U LLC. This operating system embodies the AI4U doctrine: pick the platform first, write for that platform second, verify proof third, persist the artifact fourth, record what worked and failed fifth, improve the next route sixth.
Supabase Integration
Persistent V4 Intelligence Layer
Not Configured
Not signed in
Connection Settings
Sync Actions
Authentication
Sign in to activate the Persistent V4 Ledger. Records survive browser clears, device changes, and session ends.
Persistence Status
• Local V3 Ledger: always active (localStorage)
• Persistent V4: requires Supabase config + sign in
• Offline: writes queue locally, syncs when reconnected
• RLS: each record bound to your auth.uid() only
Tech Stack — V4
Static HTML
Vanilla JS
Supabase Auth
Supabase Postgres
RLS Enforced
Vercel Deploy
GitHub Source
localStorage Fallback
Self-Improvement Engine
Learning Intelligence
Computed from your outcome logs. No fake numbers — shows empty state if no data exists.
—
Total Outcomes
—
Win Rate %
—
Top Success Platform
—
Most Failed Platform
No data yet — log outcomes to populate Learning Intelligence.