Dayflow is a private, local-first Mac app that builds an automatic timeline of screen activity into a work journal, with optional AI analysis via local or cloud providers.
Collecting history — the radar snapshots this repo daily. The trend line appears after 3 days of data (1 so far).
What it is
Dayflow is a private, automatic work journal for Mac. It runs on macOS, builds a timeline from screen activity, and can operate with local AI or cloud providers while emphasizing privacy and local storage.
How it works
Dayflow captures lightweight screen chunks, analyzes them with a chosen AI provider, and converts the day into activity cards. It offers context-aware summaries, a daily standup view, weekly reviews, and a chat interface grounded in the timeline. It supports distraction tracking, timeline export in Markdown, and local-first storage on the Mac by default. It allows AI provider choice among local models (via Ollama or LM Studio) or via Gemini, ChatGPT, or Claude.
Getting started
Install options:
- Download: Open the Dayflow DMG from GitHub Releases and install, then grant macOS Screen & System Audio Recording permission when prompted.
- Homebrew: ```bash brew install --cask dayflow
Requirements include macOS 14+, screen & system audio recording permission, and optional: Gemini API key, Ollama, LM Studio, Codex CLI, or Claude Code.
Build from source:
```bash
git clone https://github.com/JerryZLiu/Dayflow.git
cd Dayflow
open Dayflow/Dayflow.xcodeproj
Select the Dayflow scheme in Xcode and run it.
Recent releases
Latest releases include:
- v2.0.3 Dayflow 2.0.3 (2026-07-17)
- v2.0.2 Dayflow 2.0.2 (2026-07-17)
- v2.0.1 Dayflow 2.0.1 (2026-07-17)
- v2.0.0 Dayflow 2.0.0 (2026-07-03)
- v1.14.1 Dayflow 1.14.1 (2026-06-11)
Traction
GitHub stars: 6813 Forks: 399 Open issues: 90
Behind the repo
This project is maintained by JerryZLiu and is open source with a MIT license.
Caveats
License: MIT Age: Created 2025-09-23 Last push: 2026-07-17 License details and contribution guidelines are provided in the repository.