Why it matters
This release offers developers a new tool for managing daily tasks with AI assistance, particularly for those on Apple Silicon Macs. The local-first, privacy-focused design and the ability to integrate with local AI models provide flexibility and control over data and processing.

What changed

Daily, an AI-powered task manager, has been released for Apple Silicon Macs. This application is built with a focus on productivity, minimalism, and aesthetics, aiming to streamline day-to-day planning. A core feature is its day-first workspace, where tasks are organized by day rather than an endless backlog. Users can plan on a calendar-linked board, move tasks through stages like 'active,' 'discarded,' and 'done,' and embed rich content within tasks, including Markdown, code blocks, tables, tags, files, estimates, and logged time.

Daily includes a three-column board for task management, a Markdown editor with syntax highlighting and slash commands, and features like projects, tags, time logging, and soft deletion with recovery. It also maintains an activity history to track meaningful task changes. The application supports light, dark, and system appearance modes with configurable accent colors.

A significant component is the built-in task agent runtime. Unlike a simple chat widget, Daily's agent operates in a loop, streams responses, executes registered task tools, preserves conversations, and requires explicit confirmation for destructive actions. This agent can perform operations related to tasks, projects, tags, time, attachments, and summaries. It does not claim to control arbitrary applications or files on the Mac.

Users have the flexibility to choose their AI model: they can use downloadable local models for on-device inference or connect to an OpenAI-compatible provider. Daily manages the local model runtime and downloads, though these models require disk space and memory.

Privacy is a key consideration, with task data and conversation history stored locally in SQLite databases. Attachments and downloaded local models are stored as regular files. Data only leaves the Mac when a feature explicitly requires it, such as sending prompts to a configured remote AI provider or using optional sync features.

Installation can be done via Homebrew (brew install --cask scheron/tap/daily) or through a manual download of the Apple Silicon .dmg file from the project's releases. For first-time launches on macOS, a command (xattr -rd com.apple.quarantine /Applications/Daily.app) may be necessary to bypass security restrictions.

Optional sync capabilities are available, with data stored locally in SQLite. Sync options include iCloud Drive or a self-hosted sync server. The self-hosted option allows multiple Macs to sync without relying on iCloud. It can be installed on a VPS using Docker and requires minimal configuration. The first Mac to set up the server becomes the 'Parent,' controlling access for 'Child' Macs.

Why it matters for builders

Daily provides developers with a robust, local-first task management solution that integrates AI capabilities directly. The ability to run local models offers enhanced privacy and control over AI processing, which is crucial for sensitive workflows. Furthermore, the open-source nature and the self-hosted sync option cater to users who prioritize data ownership and customizability.

Practical impact

Developers can leverage Daily to organize their personal and project-related tasks with AI assistance. The integration with local AI models means builders can experiment with AI-driven task automation without sending data to external services. The self-hosted sync server option is particularly valuable for teams or individuals needing to synchronize tasks across multiple machines while maintaining full control over their data infrastructure.

Caveats and source limits

Daily is currently limited to macOS on Apple Silicon. Desktop builds for Windows and Linux are not yet available or tested. The use of local AI models requires sufficient disk space and memory, with requirements varying by model. The source does not provide specific details on the performance benchmarks of the task agent or the AI models it supports, nor does it specify pricing for any potential premium features or cloud integrations beyond the OpenAI-compatible provider option.

Sources

Written with AI assistance from the linked sources; every claim below was checked against them automatically. How we produce articles.

Claim check: 12/12 supported claims - 12 evidence links - 100% avg confidence
  • Daily is an AI-powered task manager focused on productivity, minimalism, and aesthetics.supported - github.com
  • Daily is a local-first, day-first task manager for Apple Silicon Macs.supported - github.com
  • Daily features Markdown tasks, SQLite storage, optional sync, and a built-in task agent.supported - github.com
  • Daily has a built-in task-agent runtime that streams responses, executes registered task tools, preserves conversations, and gates destructive actions.supported - github.com
  • The task agent works through Daily's own task, project, tag, time, attachment, and summary operations.supported - github.com
  • Users can choose to use a downloadable local model for inference on their Mac or connect an OpenAI-compatible provider.supported - github.com
  • Task data and conversation history are stored locally in SQLite, and attachments/local models are regular files on disk.supported - github.com
  • Data leaves the Mac only when a configured remote AI provider receives prompts, or for optional iCloud Drive or self-hosted sync.supported - github.com
  • Daily can be installed via Homebrew (`brew install --cask scheron/tap/daily`) or manual install from a `.dmg` file.supported - github.com
  • Daily ships a self-hosted sync server that installs onto any VPS running Docker.supported - github.com
  • The desktop app is supported on macOS on Apple Silicon only.supported - github.com
  • Desktop builds for Windows/Linux are not shipped or tested.supported - github.com

Caveats

  • Single-source caution: verify critical details at the linked source.
Radar score 86/100 - how it was calculated
Reliability82
Freshness92
Novelty77
Technical81
Developer96
Ecosystem72
Confidence100
  • Reliability 82: GitHub metadata supports source trust
  • Freshness 92: Fresh GitHub activity
  • Novelty 77: Fresh GitHub release
  • Technical 81: Repository technical metadata
  • Developer 96: Developer tooling signals
  • Ecosystem 72: Fresh GitHub release
  • Confidence 100: Claims have reliable evidence
Share
XLinkedInHacker News

Related articles

AI Tools - Sep 29, 2026wickrunAI: Desktop App for Orchestrating Multiple AI ModelswickrunAI is a new open-source desktop application designed to orchestrate multiple AI models, including open-source, closed-source, paid, and free options. It features a "bring your own key" (BYOK) system, ensuring API keys remain on the user's machine, and supports automatic task handover between models if one fails mid-task.AI Tools - Sep 29, 2026fongap-labs releases Delta, a local-first AI agentfongap-labs has introduced Delta, a local-first personal AI agent designed for productivity, research, and content creation. The agent prioritizes user control, recoverability, and verifiability of work processes.AI Tools - Sep 29, 2026DSH Skill Trace: Agent Skill Loading Visibility PluginPolinniZhong has released DSH Skill Trace, a local-first plugin for DeepSeek Harness that enhances visibility into which skills an agent loads during a conversation. The plugin provides detailed receipts and flow maps of skill invocations, allowing users to understand, review, and learn from the agent's operational process.AI Tools - Sep 29, 2026AgenticOS: Open-Source Platform for Building and Managing AI AgentsAgenticOS is a new open-source, self-hosted platform designed for building, running, and governing AI agents within an organization. It provides a unified environment for managing agent skills, context files, automations, and budgets, with a focus on auditability and control.AI Tools - Sep 29, 2026OpenSider for VS Code Integrates Multiple AI AgentsOpenSider for VS Code is a new open-source extension that allows developers to use multiple AI agent CLIs, such as Claude Code, Codex, Cursor, OpenCode, and GitHub Copilot CLI, from a single VS Code side panel. It offers a unified interface for these agents without bundling any models itself.AI Tools - Sep 29, 2026Jarvis AI Agent: Self-Hosted Linux Automation with Multi-LLM SupportJarvis is a self-hosted, autonomous AI agent for Linux that can control the desktop via VNC, integrate with WhatsApp, and utilize a RAG knowledge base. It supports multiple LLMs, including local Ollama models, and features a sandboxed security layer for multi-user environments.