Learning unfamiliar Mac software
Users who want spoken, screen-aware guidance and visual pointing while learning an application without repeatedly switching to a chatbot.
Independent tool overview
HeyClicky (formerly Clicky) at a glance
Clicky has evolved into HeyClicky, an active macOS voice assistant that can answer questions about the visible screen, draw guidance over apps, dictate text and run limited computer-use agents.
Visit the official HeyClicky (formerly Clicky) site ↗
- Current name
- HeyClicky
- Former name
- Clicky
- Platform
- macOS Sonoma 14.2 or later
- Windows
- Waitlist only
- Core modes
- Talk, dictation and agents
- Screen capture
- Triggered by the user hotkey
- Processing
- Cloud-based through HeyClicky and AI providers
- Starting price
- $0 with monthly limits
- Current source code
- Private; original Clicky code remains MIT-licensed
Overview
What HeyClicky (formerly Clicky) is
The project launched as an open-source AI teacher named Clicky and now continues as HeyClicky, a more capable commercial Mac app. Pressing its hotkey captures screen context and voice input so the assistant can explain what is visible, point or draw on the screen, dictate into other apps and, in Agent mode, carry out tasks rather than only describe them.
The original April 2026 Clicky code remains public under the MIT license, but its creator says subsequent product development is private. Current HeyClicky therefore should not be described as fully open source. It is a cloud-processed macOS service with a free tier and $20 and $100 monthly plans, not merely the original do-it-yourself Swift project.
Use cases
Who HeyClicky (formerly Clicky) is best for
The strongest fit depends on the job you need the product to complete, not the size of its feature list.
Hands-free questions
People who prefer holding a shortcut and asking about the current screen instead of typing and attaching screenshots manually.
Cross-app dictation
Mac users drafting messages, prompts and documents by voice in browsers, communication tools, editors and other text fields.
Guided walkthroughs
Tasks that benefit from a sequence of on-screen explanations, highlights and pauses for the user to complete manual steps.
Low-risk desktop chores
Users testing agents on reversible, well-scoped computer tasks with confirmations and clear boundaries.
Capabilities
Core HeyClicky (formerly Clicky) features
Screen-aware voice help
A hotkey sends the current screen and spoken request for cloud processing, then returns an explanation without requiring a manual screenshot upload.
On-screen pointing and drawing
Can move its own visual pointer and draw annotations over the screen to identify controls or explain a sequence spatially.
Talk mode
Provides conversational help about any visible application, with the paid plans advertising unlimited talk subject to service policies.
Agent mode
Runs computer-use tasks for the user, with a separate monthly agent-message allowance and current support for longer walkthroughs.
System-wide dictation
Streams speech into text fields across many Mac apps, cleans grammar and spelling and supports more than 50 languages according to the changelog.
Whole-document context
Can read an attached PDF, file or web page beyond only the portion currently visible on screen.
Skills
Reusable behavior or writing instructions can shape recurring work, with an early team version supporting privately shared team Skills.
Automatic model routing
HeyClicky chooses faster or deeper cloud models based on the apparent request rather than asking the user to select a provider each time.
Process
How the HeyClicky (formerly Clicky) workflow works
Step 1
Install on a supported Mac
Use macOS Sonoma 14.2 or later and grant only the microphone, accessibility and screen-recording permissions required for the intended features.
Step 2
Choose the safest mode
Use Talk for explanation, dictation for text insertion and Agent mode only when the task truly requires actions on the computer.
Step 3
Prepare the screen
Close private messages, credentials, customer data and unrelated windows before triggering a screen-aware request.
Step 4
Ask one concrete question
Press the hotkey and describe the goal, current application and preferred amount of guidance; request a walkthrough when the task has multiple steps.
Step 5
Keep approvals scoped
Review every agent confirmation and avoid blanket approval for tasks touching accounts, payments, production systems, private files or irreversible changes.
Step 6
Verify the result
Check inserted text, changed settings, sent content and completed actions yourself, especially after a long walkthrough or model-routed task.
Step 7
Clear data when appropriate
Review retained summaries and account data, then use account deletion or contact support when the information should no longer be stored.
Cost
HeyClicky (formerly Clicky) pricing and free plan
HeyClicky has a free plan plus Pro at $20 per month and Max at $100 per month. Its page also displays a 20% annual-billing option. Talk, dictation and agent messages have different allowances, so the agent quota—not the word 'unlimited'—is the important limit for autonomous work.
Free
$0
Free-forever trial tier with no card required.
- 25 talk messages per month
- Limited dictation described as 'plenty' rather than a numeric allowance
- 25 agent messages per month
- macOS Sonoma 14.2 or later
Pro
$20/month
Everyday-use tier with unlimited talk and dictation plus a larger agent allowance.
- Unlimited talk
- Unlimited dictation
- 150 agent messages per month
- 20% annual option shown on the pricing page
Max
$100/month
Power-user tier centered on substantially more agent work.
- Unlimited talk
- Unlimited dictation
- 1,000 agent messages per month
- 20% annual option shown on the pricing page
Teams pilot
Contact HeyClicky
Early manually onboarded team version with mixed Pro and Max seats.
- Central member and billing dashboard
- Private team-shared Skills
- Described as an early v0 rather than a standard self-serve plan
- Confirm current availability and controls before rollout
Pricing checked . Check current pricing at the source ↗
Assessment
HeyClicky (formerly Clicky) strengths and limitations
Where it stands out
- The cursor-adjacent, screen-aware interface reduces context switching while learning unfamiliar software.
- Spoken explanations plus visible pointing can be more usable than a text-only answer for interface questions.
- Talk, dictation and agent modes cover explanation, text entry and action without requiring separate apps.
- Whole-document reading extends the product beyond whatever happens to fit on the current screen.
- The free tier is enough to test the interaction model before paying.
- Recent changelog entries show frequent reliability, accessibility and privacy changes rather than an abandoned prototype.
- The original MIT-licensed Clicky code remains available for developers who prefer a do-it-yourself base.
What to consider
- HeyClicky is currently Mac-only and requires macOS Sonoma 14.2 or later; Windows remains a waitlist.
- The current product's new development is private even though the original Clicky codebase remains open under MIT.
- Screen, audio, prompts and transcripts are processed in the cloud and forwarded to providers including Anthropic, OpenAI, Deepgram and Cerebras.
- The company says raw screenshots are not stored, but it retains the AI-generated screen analysis and the user's prompt to provide context.
- The privacy policy uses a purpose-based retention period rather than publishing a short fixed deletion schedule for ordinary account data.
- HeyClicky has not obtained SOC 2 or ISO 27001 certification according to its trust page.
- Screen-recording and accessibility permissions are powerful; a compromised account, app or agent could expose information or take unintended actions.
- The new 'Always approve' option reduces interruptions but also removes a valuable checkpoint during long agent tasks.
- The free tier includes only 25 talk and 25 agent messages per month, while Pro still caps agents at 150 messages.
- Automatic model routing makes the exact provider and behavior less predictable from one request to another.
- Visual pointing can identify the wrong control, and agents can misunderstand state changes that occur after a screenshot or earlier step.
- There is no published enterprise certification package, formal availability commitment or standard team pricing on the public site.
Compare
HeyClicky (formerly Clicky) alternatives
The right alternative depends on the specific output, workflow, controls and budget your project requires.
Agents
Claude Cowork
Choose Claude Cowork for Anthropic's broader desktop-agent workflow and file work rather than a cursor-first teaching interface.
Explore Claude Cowork →Consumer
Gemini for Mac
Choose Gemini for Mac for a native Mac client tied directly to Google's Gemini ecosystem.
Explore Gemini for Mac →Business Operations
ChatGPT
Choose ChatGPT for a mature general assistant with desktop voice, vision and broader cross-platform access.
Explore ChatGPT →Agents
OpenClaw
Choose OpenClaw for messaging-driven autonomous tasks and a more developer-operated agent stack.
Explore OpenClaw →Questions
HeyClicky (formerly Clicky) FAQs
Is Clicky still active?
Yes. The product is now called HeyClicky and continues to receive frequent Mac app updates. The original Clicky code remains available separately.
Is HeyClicky open source?
Not fully. The original April 2026 Clicky codebase is public under the MIT license, but its creator says newer HeyClicky development is private.
Is HeyClicky free?
It has a free tier with 25 talk messages and 25 agent messages per month plus a limited dictation allowance. Pro is $20 monthly and Max is $100 monthly.
Does HeyClicky watch my screen continuously?
The company says no. Screen capture occurs when the user presses the hotkey, and its trust page says background collection does not include screen, files, clipboard or keystrokes.
Are HeyClicky screenshots stored?
HeyClicky says raw screenshots are not stored after processing. It does retain the AI-generated screenshot analysis and the user's prompt for context.
Is processing local?
No. The current service processes requests in the cloud and forwards relevant screen, voice, transcript or prompt content through its backend to third-party AI providers.
What is the difference between Talk and Agents?
Talk answers and guides the user, while Agent mode performs tasks. Agent messages have a separate monthly quota even on paid plans.
Can HeyClicky type into other apps?
Yes. Its dictation feature is designed to insert speech as cleaned-up text across Mac text fields, and screen-aware prompts can draft based on visible context.
Does HeyClicky work on Windows?
The official product does not yet. The site lists macOS Sonoma 14.2 or later and offers a Windows waitlist. Community-built Windows ports are separate projects.
Should I let HeyClicky agents always approve actions?
Use blanket approval only in tightly scoped, reversible environments. Keep confirmations enabled for email, accounts, payments, production systems, private files and any destructive or externally visible action.
Bottom line
Our HeyClicky (formerly Clicky) verdict
HeyClicky has grown from a delightful open-source teaching demo into a distinctive Mac assistant that combines screen-aware voice help, dictation and computer-use agents. The interface is its real advantage. The tradeoff is trust: powerful Mac permissions, cloud processing, retained text analysis, private current development and no SOC 2 or ISO certification. It is worth testing on the free tier with low-risk tasks and tightly scoped approvals before giving Agent mode broader access.
Visit HeyClicky (formerly Clicky) website ↗Similar Consumer Tools
View all Consumer AI tools →
Muse Spark
Muse Spark - Meta Superintelligence Labs' new multimodal reasoning model
Claude Opus 4.7
Claude Opus 4.7 - Anthropic’s new top model with advanced agentic coding
GLM-5.1
GLM-5.1 — Z AI's new open-source flagship model built for long-horizon agentic coding
Kimi K2.6
⚙️ Kimi K2.6 - Moonshot AI's powerful open-source coding and agent model
AI training for the future of work.
Get access to all our AI courses, hundreds of real-world AI use cases, live expert-led workshops, an exclusive network of AI early adopters, and more.
AI Courses
Get unlimited access to all of our current & upcoming industry-specific AI courses for the duration of your subscription.
Daily Guides
To keep up with the rapid pace of AI, our team publishes AI implementation guides daily. Our library contains 300+ practical use cases to automate real-world work.
Workshops
Join weekly, live, interactive sessions with industry leaders who are at the forefront of AI for hands-on implementation guidance and exclusive insights.
Community
Network with an exclusive community of AI-first professionals who are working smarter with AI. Learn how early adopters are using AI in their work and businesses.