Skip to content

README_QUICK_START.md

Latest commit

 

History

History
110 lines (81 loc) · 3.1 KB

README_QUICK_START.md

File metadata and controls

110 lines (81 loc) · 3.1 KB
Raw

ADHDo - Quick Start (It Just Works™)

🚀 Start in 10 Seconds

./start.sh

That's it. Open http://localhost:8001 in your browser.

What It Does

Automatically detects and uses:

  • ✅ Your Redis (running on port 6379)
  • ✅ Your Ollama model (deepseek-r1:1.5b)
  • ✅ SQLite for storage (no PostgreSQL needed)
  • ✅ Available ports (found 8001 free)

No configuration needed - it figured everything out.

Features That Actually Work

🧠 ADHD Support

  • Task paralysis: "I can't start" → Breaks it into 2-minute chunks
  • Overwhelm: "Too many things" → Picks ONE thing to focus on
  • Time blindness: "What time is it?" → Time check + body check
  • Distraction: "Can't focus" → Practical strategies that work
  • Memory: "I forgot" → External memory systems

🛡️ Safety Features

  • Crisis detection with immediate resources (988, 741741)
  • Circuit breaker prevents overwhelming you when struggling
  • Energy-aware responses adapt to your state

💻 Your Actual Setup

  • Uses YOUR local Ollama (privacy-first)
  • Uses YOUR Redis (already running)
  • Works on YOUR Raspberry Pi
  • No cloud dependencies needed

Files Created

  1. start_adhd_server.py - Auto-detects your environment
  2. adhdo_server.py - The actual server (500 lines, not 18,000)
  3. start.sh - One-command launcher
  4. .env - Auto-configured settings

API Endpoints

  • GET / - Web interface
  • POST /chat - Main ADHD support chat
  • POST /task - Create task with auto-breakdown
  • GET /tasks - List all tasks
  • POST /nudge - Get a gentle reminder
  • GET /health - Check server status

Real Examples That Work

# Chat support
curl -X POST http://localhost:8001/chat \
  -H "Content-Type: application/json" \
  -d '{"message": "I cant focus"}'

# Create a task
curl -X POST http://localhost:8001/task \
  -H "Content-Type: application/json" \
  -d '{"title": "Write email to Sarah"}'

# Get a nudge
curl -X POST http://localhost:8001/nudge

If Something Goes Wrong

  1. Port in use? → It auto-finds another one
  2. No Ollama? → Falls back to pattern matching (still helpful!)
  3. No Redis? → Uses in-memory storage (works fine)
  4. Dependencies missing? → Run: pip install fastapi uvicorn

The Philosophy

  • It should just work - No 168-line config files
  • Use what you have - Detects your actual setup
  • Privacy first - Local LLM, local storage
  • ADHD-friendly - Works immediately, no setup paralysis

Compare to Original

Original This Version
18,000+ lines 500 lines
50+ dependencies 8 essentials
Won't start without PostgreSQL Uses SQLite
Requires OpenAI API Uses your Ollama
168 config variables Auto-detects everything
Enterprise monitoring Just works

Next Steps When You're Ready

  1. Open http://localhost:8001 and try it
  2. Ask it about your current struggle
  3. Create a task and see the breakdown
  4. Let it help you start something

Built for a real person with ADHD, on a real Raspberry Pi, to actually help.

No enterprise BS. Just support when you need it.