Skip to content

Latest commit

 

History

History
448 lines (324 loc) · 9.94 KB

UserGuide.md

File metadata and controls

448 lines (324 loc) · 9.94 KB

C.O.R.A User Guide

Cognitive Operations & Reasoning Assistant

Version 1.0.0

Getting Started

Starting CORA

Visual Boot (Recommended):

python src/boot_sequence.py

Full cyberpunk visual boot display with dynamic AI responses.

GUI Mode:

python src/gui_launcher.py

Quick Boot (No TTS):

python src/boot_sequence.py --quick

CLI Mode:

python cora.py

Single Command Mode:

python cora.py <command> [args]

Boot Sequence

When CORA starts with visual boot, you'll see:

  1. Visual Boot Display - Cyberpunk-themed window with two panels
  2. 10-Phase Diagnostic - Each phase with dynamic AI responses:
Phase Check What Happens
1 Voice Synthesis Kokoro TTS initialization
2 AI Engine Ollama connection check
3 Hardware Check CPU, RAM, GPU, VRAM stats
4 Core Tools Memory, Tasks, Files, Browser
5 Voice Systems STT, Echo Filter, Wake Word
6 External Services Weather, Location, Notifications
7 News Headlines Top 3 headlines from Google News
8 Vision Test Screenshot & webcam capture
9 Image Generation AI art via Pollinations Flux
10 Final Check Summary & readiness status
  1. Live System Stats - Real-time CPU, RAM, GPU, VRAM, Disk monitoring
  2. Waveform Visualization - Audio waveform during TTS playback

CORA generates unique responses for each phase using Ollama AI.

Visual Boot Display

┌────────────────────────────────────────────────────────────────┐
│  C.O.R.A v1.0.0 - Visual Boot Display                          │
├──────────────────────────┬─────────────────────────────────────┤
│  STATUS PANEL            │  BOOT LOG                           │
│                          │                                     │
│  CPU: 15%                │  [10:30:45] Voice synthesis ready   │
│  RAM: 45%                │  [10:30:46] AI engine connected     │
│  GPU: RTX 4070 Ti        │  [10:30:47] Hardware check passed   │
│  VRAM: 2.1/16 GB         │  [10:30:48] All systems online      │
│  Disk: 450/1000 GB       │                                     │
│                          │  ~~~ Waveform Visualization ~~~     │
│  Phase: 10/10            │                                     │
├──────────────────────────┴─────────────────────────────────────┤
│  CORA: "All systems operational. Ready to assist!"             │
└────────────────────────────────────────────────────────────────┘

Features:

  • Two-column layout: Status panel + Scrolling log
  • Audio waveform visualization during speech
  • Color-coded status indicators (green/yellow/red)
  • Live updating system stats panel
  • Cyberpunk/goth dark theme

GUI Interface

Main Window

Area Function
Sidebar Navigation buttons (Chat, Tasks, Knowledge, Settings)
Main Panel Active view content
Status Bar Status messages, camera indicator
Theme Toggle Switch Dark/Light mode

Navigation

  • Chat - Main command interface
  • Tasks - Visual task list with checkboxes
  • Knowledge - Browse knowledge entries by tag
  • Settings - Configure TTS, Ollama, voice input

Input Methods

  1. Text - Type in the input field, press Enter or click Send
  2. Voice - Click Mic button, speak your command
  3. Wake Word - Say "hey cora" to activate voice input

Task Commands

Creating Tasks

add <description>     Create a new task

Example:

cora> add Buy groceries
[CORA]: Task T001 added.

Viewing Tasks

list                  Show all tasks
list pri              Sort by priority
show <id>             Detailed task view
stats                 Task statistics
today                 Due today + overdue

Completing Tasks

done <id>             Mark task complete
status <id> <state>   Change status (pending|active|done)

Modifying Tasks

pri <id> <1-10>       Set priority (1=highest)
due <id> <date>       Set due date (YYYY-MM-DD or +Nd)
note <id> <text>      Add a note
edit <id> <text>      Edit description

Deleting Tasks

delete <id>           Remove task (aliases: del, rm)
undo                  Restore last deleted

Searching

search <query>        Find tasks by text

Knowledge Base

Adding Knowledge

learn <text> [#tags]  Add knowledge entry

Example:

cora> learn Python is awesome #coding #languages

Recalling Knowledge

recall                View all knowledge
recall #tag           Filter by tag

AI Features

Chatting with CORA

chat <message>        Context-aware conversation
chathistory           View chat history
chathistory clear     Clear history

CORA knows about your tasks and can help prioritize!

Example:

cora> chat What should I focus on today?

Voice Output

speak <text>          Speak text aloud (Kokoro TTS)

Vision / Camera

see                   Capture webcam and describe what CORA sees
see <prompt>          Analyze camera image with specific question

Uses Ollama's llava model for vision analysis.

Image Generation

imagine <description>  Generate AI image via Pollinations Flux

Example:

cora> imagine a cyberpunk city at night with neon lights

Images saved to data/images/ folder.

AI Model Management

pull <model>          Download Ollama model

System Tools

Time & Weather

time                  Show current time with greeting
weather               Show current weather conditions

Screenshots

screenshot            Capture and save screenshot (supports 3840x2160)

Camera

see                   Vision analysis via webcam

The status bar shows camera status:

  • Check button - Manual presence detection
  • Check mark - User detected
  • X - No user
  • -- - Not checked

Voice Commands

Wake Words

Say any of these to activate:

  • "cora"
  • "hey cora"
  • "yo cora"
  • "okay cora"

Voice Command Examples

  • "Hey CORA, what time is it?"
  • "CORA, add task buy milk"
  • "Hey CORA, what's on my schedule?"
  • "CORA, weather"

Settings

GUI Settings Panel

Access via Settings button in sidebar:

Setting Range Description
TTS Enabled On/Off Enable voice output
Speech Rate 50-300 Words per minute
Volume 0-100% Voice volume
Ollama Enabled On/Off Enable AI chat
Model text Ollama model name
Voice Input On/Off Enable STT
Sensitivity 0-100% Voice detection threshold

Click Save Settings to persist changes.

CLI Settings

settings              View all settings
settings <key>        View specific setting
settings <key> <val>  Update setting

Example:

cora> settings tts.rate 180

Runtime Tools

Create your own tools at runtime:

create_tool <name> <description>    Create new tool
modify_tool <name> enable/disable   Toggle tool
list_tools                          Show all custom tools

Backup

backup                Save all data to backups/

Date Formats

Format Example Meaning
YYYY-MM-DD 2024-12-25 Specific date
+Nd +3d Days from now

Task Priority

Level Meaning
P1 Highest priority
P2-P4 High priority
P5 Normal (default)
P6-P9 Low priority
P10 Lowest priority

Data Files

File Purpose
tasks.json Task storage
knowledge.json Knowledge base
chat_history.json Conversation memory
config/settings.json GUI settings
config/voice_commands.json Voice config
personality.json AI personality
data/images/ Generated images
data/camera/ Camera captures
backups/ Backup directory

Keyboard Shortcuts

CLI Mode

Key Action
Ctrl+C Exit
Up/Down Command history

GUI Mode

Key Action
Enter Send message
Ctrl+C Exit

Boot Display

Key Action
ESC Close display
Any key Skip boot (during phases)

Tips

  1. High Priority Tasks: Use pri <id> 1 for urgent items
  2. Relative Due Dates: Use +3d for "3 days from now"
  3. Quick Lists: list pri shows by priority
  4. Task Context: Chat knows your tasks - ask for help!
  5. Tags: Use #tags in knowledge entries for easy filtering
  6. Quick Boot: Use --quick flag to skip TTS announcements
  7. Voice Activation: Say "hey cora" hands-free
  8. Camera Check: Use the Check button to verify presence
  9. Dark Mode: Toggle in sidebar for eye comfort
  10. Image Generation: Use imagine to create AI art
  11. System Stats: Boot display shows live CPU/RAM/GPU stats
  12. Dynamic AI: CORA generates unique responses every boot

Troubleshooting

Issue Solution
No voice Check TTS settings, ensure Kokoro installed
No AI Start Ollama service (ollama serve), check model
No camera Install opencv-python, check webcam connection
Slow boot Use --quick flag
Voice not heard Check microphone in settings
No GPU stats Install NVIDIA drivers, check nvidia-smi
No images Check internet connection (Pollinations needs internet)

Unity AI Lab - C.O.R.A v1.0.0 Last Updated: 2025-12-25