AI can make mistakes. Information may be incomplete, outdated or incorrect. Always verify important details from reliable sources before making decisions.
list_alt Table of Contents
flag Overview
DOONKLY is a Chrome extension that allows you to interact with different AI models directly from your browser. Each icon in the toolbar activates a specific functionality.
Icon Legend:
- screenshot_region Screenshot - Capture page area
- rocket_launch Rocket - Ultra-fast responses
- school School - University/scientific mode
- network_intelligence_history Advanced - Deep reasoning
- palette Palette - Image generation
- attach_file_add Upload - Upload files
- mic Voice - Audio recording
| Mode |
Speed |
Ideal For |
| rocket_launch Rocket |
electric_boltelectric_boltelectric_boltelectric_boltelectric_bolt |
Simple questions, translations |
| send Standard |
electric_boltelectric_boltelectric_bolt |
General conversations |
| network_intelligence_history Advanced |
electric_boltelectric_bolt |
Deep analysis |
| school University |
electric_boltelectric_bolt |
Mathematics, sciences |
| palette Palette |
electric_bolt |
Image generation |
palette Image Creation
Model used: DALL-E 3
Resolution: 1792x1024 (16:9, HD quality)
How to Use
- Click on the palette icon (palette)
- Write your prompt with detailed description
- Press Send
- View the result with Expand and Download options
EXAMPLE PROMPT:
palette
"A space cat wearing an astronaut suit, floating among the stars with colorful planets in the background, modern cartoon style"
send
Tips:
- done_all Use detailed and specific descriptions
- done_all Specify the desired artistic style
- done_all Indicate colors, atmosphere, and mood
- cancel Avoid vague prompts
- cancel Don't request text in images
rocket_launch Rocket Mode
Model used: gpt-5-nano
Ideal for: Short answers, quick translations, simple questions
EXAMPLE PROMPT:
rocket_launch
"Analyze the text (printscreen) and convert to english'"
send
Use Cases:
- Quick translations
- Unit conversions
- Quick definitions
- Simple questions
Limitations:
- cancel Not optimal for complex analysis
- cancel Limited context (max 2000 output tokens)
school University Mode (School)
Model used: GPT-4 Turbo
Ideal for: Mathematics, physics, chemistry, scientific research, formulas
EXAMPLE PROMPT:
school
"Explain the derivative of x³ + 2x² - 5x + 1 step by step"
send
Special Features:
- done_all MathJax rendering automatic for formulas
- done_all Temperature 0.1 for maximum precision
- done_all Detailed step-by-step explanations
- done_all Export to SVG and TXT
- cancel PDF export disabled (complex formulas)
network_intelligence_history Think More Mode (Advanced)
Model used: GPT-4 Turbo
Ideal for: Deep analysis, complex reasoning, articulated problem-solving
EXAMPLE PROMPT:
network_intelligence_history
"Analyze the pros and cons of implementing an AI system in a company of 100 employees"
send
When to Use:
- done_all Complex business decisions
- done_all Multi-factorial analysis
- done_all Complex system design
- done_all Ethical or philosophical reasoning
screenshot_region Screenshot Area
Function: Captures a selected portion of the current web page
Format: High-quality PNG
Auto-confirm: 1 second after selection
How to Use
- Click the Screenshot icon (screenshot_region)
- Click and drag to create a selection rectangle
- Wait 1 second for auto-confirmation
- Add question about the image and press Send
EXAMPLE PROMPT:
screenshot_region
"Analyze the data in this table and give me a summary of the main trends"
send
Use Cases:
- Document analysis
- Code debugging
- Design feedback
- Error troubleshooting
attach_file_add Upload Files
Supported formats: Text files (.txt, .md, .json, .csv), Images (.jpg, .png, .gif, .webp), Code files (.js, .py, .java, .cpp)
How to Use
- Click the Upload icon (attach_file_add)
- Select one or more files
- Files appear in container above textarea
- Write question and press Send
EXAMPLE PROMPT:
attach_file_add
"Find bugs in these files and suggest fixes"
send
Use Cases:
- Code review
- Data analysis
- Text processing
- Version comparison
mic Voice Recording
Function: Records audio from computer microphone
Duration: ~10 seconds max recommended
Format: WebM/Opus
How to Use
- Click the Voice icon (mic)
- Allow microphone permission (first time)
- Speak into the microphone
- Click again to stop recording
- Press Send (with or without text)
Use Cases:
- Quick hands-free questions
- Text dictation
- Foreign language pronunciation
Limitations:
- warning 10 seconds max recommended
- warning Requires browser microphone permission
- warning Audio quality depends on microphone
chat Standard Mode (GPT-4o)
Model used: GPT-4o
Ideal for: General conversations, various questions, daily use
When no icon is selected, pressing Send automatically uses GPT-4o.
EXAMPLE PROMPT:
"Explain in simple words what blockchain is"
send
Advantages:
- done_all Medium speed
- done_all Versatile for many question types
- done_all Maintains conversational context
- done_all Supports images and files
- done_all Balance creativity/precision
lightbulb Tips & Tricks
Choose the Right Mode
cancel Not optimal: Use University mode for "What is 2+2?"
check Better: Use Rocket mode for simple questions
Improve Response Quality
cancel Vague: "Tell me about Python"
check Specific: "What are the differences between list and tuple in Python, with practical examples?"
Use Screenshots for Visual Context
Instead of: "There's an error in my code"
Better: [Screenshot of code] + "Why am I getting this error on line 23?"