Dayflow
The automatic work journal. Privately turns your screen into a timeline of what you actually accomplished. Open-source and local-first.
Last verified:
What is Dayflow?
Dayflow is a native macOS application that automatically tracks your screen activity and generates a detailed visual timeline of your day with AI-powered summaries. It records your screen at 1 frame per second and analyzes activity every 15 minutes using AI to understand context—for example, distinguishing between "researching on YouTube" versus "watching cat videos." Unlike traditional time trackers that only log which apps are open, Dayflow understands what you're actually doing, creating an accurate automated log without manual tagging.
Key features include automatic activity timelines with AI-generated summaries, watchable timelapses of your day, distraction highlights to identify what pulled you off-task, and the ability to export timelines as Markdown for any date range. The app is extremely lightweight at only 25MB, uses approximately 100MB of RAM, and consumes less than 1% CPU. Dayflow offers three AI processing options: Gemini (bring your own API key), local models via Ollama or LM Studio for maximum privacy, or ChatGPT/Claude for best quality.
Dayflow is designed for developers, freelancers, founders, students, AI researchers, and knowledge workers who want a truthful, automated log of their day without spending time categorizing activities themselves. It is free and open-source under an MIT license, meaning you can inspect the code, audit data handling, and remix it however you like. The app requires macOS 13.0 (Ventura) or later and is built with SwiftUI for native performance.
Privacy is a core design principle—when using Local Mode with Ollama or LM Studio, all data stays completely on your device and never leaves your Mac. Recordings and databases are always stored locally at ~/Library/Application Support/Dayflow/recordings/. You have full control to start/stop recording from the menu bar, delete recordings at any time, and configure privacy settings based on your AI provider choice.
Dayflow pricing
Pricing model: Freemium
Dayflow is free and open-source (MIT Licensed) for personal use. The Personal plan is completely free and powered by advanced AI that runs 100% locally. It includes 100% local & private processing, automatic activity timeline, AI-powered summaries, bring your own API keys, open source code, and no account required. You can use it for free with Local models (Ollama/LM Studio) or Gemini's free tier. Using ChatGPT or Claude requires their respective paid subscriptions (Plus/Pro at $20+/month) to access the API via CLI. An Enterprise plan is available with SSO/SAML integration and data retention controls—contact Dayflow for enterprise pricing. The Enterprise plan includes everything in Personal plus team security features.
Dayflow pros
- Automatic timeline generation without manual tagging
- Context-aware AI distinguishes research from procrastination
- Open-source with permissive MIT license
- Extremely lightweight: 25MB app size, ~100MB RAM, <1% CPU
- 100% local processing option with Ollama/LM Studio for maximum privacy
- Visual timelapses to review how time was spent
- Distraction highlights identify context switching
- Export timelines as Markdown for any date range
- Free for personal use with local models or Gemini free tier
- Menu bar control for instant start/stop recording
- Automation support via Shortcuts and deeplinks
- Choose between Gemini, ChatGPT, Claude, or local AI providers
- All recordings stored locally on your Mac
- Native SwiftUI app optimized for macOS performance
- No account required to use
- Generate client-ready summaries for freelancers
- Verification of what really shipped for engineers
- Bring your own API keys for cloud AI providers
Dayflow cons
- macOS only—no Windows or Linux support
- Requires macOS 13.0 (Ventura) or later
- Local AI mode requires significant GPU power and battery
- Local models may be less detailed than cloud AI
- Requires Screen & System Audio Recording permissions
- ChatGPT/Claude require paid Plus/Pro subscriptions ($20+/month)
- Local AI uses ~33+ LLM calls per segment vs 2 for Gemini
- Must install Ollama or LM Studio separately for local mode
- API errors possible if Gemini API key is inactive
- May need to restart app after granting permissions
Frequently asked questions about Dayflow
What does Dayflow do?
Dayflow records your screen at 1 frame per second and uses AI to analyze your activity every 15 minutes. It builds a visual timeline with AI-generated summaries of what you were actually doing, understanding context like the difference between researching on YouTube versus watching cat videos. Unlike traditional trackers that just log app names, you don't have to manually tag anything.
Is Dayflow private and where is my data stored?
Yes, you have full control. In Local Mode using Ollama or LM Studio, your data never leaves your device. If you choose Gemini or ChatGPT/Claude, data is sent to the respective providers for analysis, but you can configure privacy settings. Recordings and databases are always stored locally on your Mac at ~/Library/Application Support/Dayflow/recordings/. Dayflow is open-source (MIT license) so you can verify exactly how your data is handled.
How much does Dayflow cost?
Dayflow is free and open-source (MIT Licensed). You can use it for free with Local models or Gemini's free tier. The Personal plan is free for personal use and includes 100% local & private processing. Using ChatGPT or Claude requires their respective paid subscriptions (Plus/Pro at $20+/month) to access the API via CLI. Enterprise pricing is available by contacting Dayflow directly.
What AI providers can I use with Dayflow?
Dayflow supports three AI processing options: Gemini (recommended for most users, get a free API key from Google AI Studio, most efficient with only 2 LLM calls per timeline segment), Local Models via Ollama or LM Studio (maximum privacy, all processing on your Mac, requires more power with ~33+ LLM calls per segment), or ChatGPT/Claude (best quality, requires paid Plus/Pro subscription, uses 4-6 LLM calls per segment with frontier reasoning models). You can switch providers at any time in settings.
What are the system requirements for Dayflow?
Dayflow is a native macOS app built with SwiftUI that requires macOS 13.0 (Ventura) or later. The app is extremely lightweight at 25MB size, uses approximately 100MB of RAM, and consumes less than 1% CPU. For local AI mode, you need to install Ollama or LM Studio separately, and it's recommended to plug in during long sessions due to GPU usage.
Who is Dayflow for?
Dayflow is for developers, freelancers, founders, students, AI researchers, and knowledge workers who want a truthful, automated log of their day without having time to categorize it themselves. Founders use it to see where days disappear and balance maker vs manager time. Engineers verify what really shipped and generate standup recaps. Students make study sessions visible and reduce distraction drift. Freelancers generate client-ready summaries and show proof of work.
Can I start and stop recording manually?
Absolutely. You can start/stop recording instantly from the menu bar by clicking the Dayflow icon and selecting Start Recording. You can also automate it using Shortcuts via deeplinks/URL schemes. You have full access to delete recordings and data from your local folder at any time.
How accurate is the AI analysis?
It depends on the model you choose. Gemini and ChatGPT/Claude generally offer the highest accuracy and narrative quality. Local models are improving rapidly but may be slightly less detailed. You can switch providers at any time in settings to find the best balance for your needs. Historical timeline data is preserved when switching providers.
What features does Dayflow include?
Dayflow includes automatic activity timeline with AI-generated summaries, watchable timelapses of your day to see exactly how time was spent, distraction highlights to identify what pulled you off-task and reduce context switching, and timeline export as Markdown for any date range to share or archive. You can view your timeline, export it, watch timelapses, and set up automation via URL schemes.
How do I install Dayflow and grant permissions?
Download the latest Dayflow.dmg from GitHub Releases or install via Homebrew with brew install --cask dayflow. Open Dayflow.dmg and drag it to your Applications folder, then launch from Applications. When prompted, grant Screen & System Audio Recording permissions by going to System Settings → Privacy & Security → Screen & System Audio Recording and enabling the checkbox next to Dayflow. You may need to restart Dayflow after granting permissions for them to take effect.