Anchor

AnchorOn-Device AI Focus Companion for macOS

Free macOS focus assistant that runs on-device AI to tell deep work from distraction and nudge attention back with reminders and pet rewards.

freemium
focusmacOSon-device AItime managementanti-distractionprivacyGitHub Pagesindie app
Indexed
Updated
3.4 (0 Number of reviews)

Log in to rate the project

Try Now

What Anchor Does

Anchor is a native macOS focus companion that uses on-device AI to interpret what is on screen, decide whether the current activity is deep work or a distraction, and pull the user back when attention drifts.

Key Features

  • Screen-context analysis instead of static app blocklists
  • Real-time reminders, spoken cues, and per-session focus scores
  • Collectible pets awarded or lost based on session performance
  • All processing local, so screenshots are not uploaded off the device

How It Is Distributed

The app ships as a free download from a personal GitHub Pages landing page maintained by developer HaichaoLihc, with builds hosted through GitHub Releases. It requires macOS 14.1 or later on Apple silicon and does not currently offer Windows or web builds.

Who It Fits

Students, researchers, and remote workers who prefer a lightweight, privacy-focused focus tool over server-based productivity suites and use Apple silicon Macs as their main device.

Pros & Cons

Pros

  • Runs AI locally, no screenshots uploaded off the device
  • Free download from GitHub Releases
  • Reads screen context instead of relying on static blocklists
  • Focus scoring and pet rewards add lightweight gamification

Cons

  • macOS only, requires Apple silicon and macOS 14.1 or newer
  • Small indie project without a large support team behind it

Frequently Asked Questions

Does Anchor need internet or cloud access?

The developer states that AI runs on-device and screenshots stay local, so it does not depend on a cloud model.

What are the system requirements?

macOS 14.1 or later on Apple silicon Macs.

How is it distributed?

As a free download from a GitHub Pages site, with binaries served through GitHub Releases.

Explore More

Similar Tools

Guage

Guage

Guage is a free Chrome extension for Claude.ai users who want clearer visibility into usage limits and an easier way to continue conversations elsewhere. It displays session usage, remaining messages, reset times, and warnings when a limit is near. When Claude stops a conversation because of a limit, Guage packages the current chat into a context summary and sends it to the input box of ChatGPT, Gemini, Grok, or DeepSeek. The extension requires no account, can save conversations, and supports Markdown export. Its scope is limited: Guage currently works on the Claude.ai website through Chrome, not on desktop or mobile clients.

Sutra

Sutra is a local-first prompt manager for people who reuse the same instructions across ChatGPT, Claude, Gemini, and Google AI Studio. It keeps prompts on the local device, requires no account, and provides folders, tags, and reusable variables for organizing templates. Users can copy rendered prompts or insert saved content directly into supported chat interfaces. The approach is practical for privacy-conscious users, writers, researchers, and anyone comparing several AI models. The tradeoff is equally clear: public documentation is limited, platform support may require verification, and there is no built-in cloud synchronization for moving a library between devices.

Bristol

Bristol

Bristol is a report-generation tool from LayerProof that turns uploaded files or data into interactive web reports through a chat-based editing workflow. Users can describe structural or content changes conversationally, then publish the result as a live webpage rather than sending another static attachment. The concept is aimed at agencies, freelancers, consultants, and other people who regularly deliver client-facing reports. Public information remains limited, however. Pricing, supported file formats, language support, privacy controls, and export options have not been clearly documented, so Bristol is best treated as a tool to watch until an official trial or fuller product documentation becomes available.

Simple PDF Wizard

Simple PDF Wizard

Simple PDF Wizard is a browser-based PDF toolkit built around privacy, speed, and a deliberately small feature set. It handles everyday jobs such as editing, merging, compressing, signing, and converting PDFs without requiring an account or adding watermarks. Its standout feature, AI-Optimizer, converts PDF content into structured Markdown intended for LLM and RAG workflows, potentially reducing cleanup work and token usage. Files are advertised as being processed locally on the device, with a 50MB limit and a free tier offering three tasks per day. It is a practical fit for occasional PDF work, sensitive documents, and users who need clean text extraction rather than a full professional document suite.

VELOCE AI

VELOCE AI

VELOCE AI is a Windows download manager built around automatic download optimization, an embedded Chromium browser, media detection, and support for up to 32 concurrent connections. It costs $5 as a one-time purchase, includes lifetime updates, and has no subscription or advertising. A three-day trial is available without a credit card. The app is aimed at users who want a simpler alternative to established tools such as IDM, especially for downloading media and managing files from within one window. Its AI acceleration claims are promising but still need independent testing, and platform support is currently centered on Windows 10 and 11, with a separate free Android version.

WithAI.Top

WithAI.Top

WithAI.Top is a free AI tool directory that organizes generative AI apps, models, and automation products by practical categories such as text, image, audio, coding, and marketing. Each listing includes a short description and relevant tags, helping visitors discover alternatives without knowing a specific model or product name in advance. The site also includes a Submit Tool entry for developers who want to add their products. WithAI.Top works best as a first-pass discovery layer rather than a replacement for official documentation, privacy reviews, pricing checks, or hands-on testing. Its straightforward layout can save time for newcomers, product researchers, indie developers, and anyone comparing AI tools for a specific task.

Open-source Alternatives

PriceAI: AI Subscription Comparison Tool Aggregating 100+ Channels

PriceAI is an open-source AI subscription comparison tool that aggregates prices from over 100 channels for services like ChatGPT, Claude, Gemini, and Grok. It displays real-time lowest prices, stock status, and direct purchase links, helping users find the most cost-effective subscription channels. The project is developed in TypeScript and had 1212 stars at the time of collection.

agent-device: Let AI Agents Control Mobile Devices via CLI

agent-device is an open-source command-line tool that empowers AI agents to directly control iOS and Android devices through a CLI interface. Built with TypeScript, it supports essential operations like taps, swipes, and text input, making it easy to integrate into automation workflows. It is ideal for developers and testers who need AI to interact with real mobile devices. The project is licensed under MIT and has 2916 GitHub stars as of collection time.

Banana Slides: AI-native slide generator built on Nano Banana Pro

Banana Slides is an AI-native slide generator built on Nano Banana Pro. It accepts a single sentence, an outline, or an uploaded document to produce editable PPTX or PDF decks with transitions, extractable text, and optional AI voiceover narration. It runs locally or in Docker under an AGPL-3.0 license, noted as non-commercial. Primary languages are Python and React. As of collection, it has 14,811 stars on GitHub.

DreamServer: Turn Your Computer into a Versatile AI Server

DreamServer is an open-source project that transforms your PC, Mac, or Linux machine into a versatile AI server. It integrates LLM inference, chat UI, voice interaction, agents, workflows, RAG, and image generation. Designed for individual developers and small teams, it runs most models without a dedicated GPU, offering a private and cost-effective AI infrastructure. The project is primarily written in Shell and licensed under Apache-2.0.

aistore: Open-source storage for large-scale AI training and inference

aistore is an open-source storage system from NVIDIA, built for large-scale AI training and inference. It offers both object storage and file system interfaces, scaling up to hundreds of petabytes, and integrates deeply with popular AI frameworks to eliminate data bottlenecks. The project is primarily written in Go and released under the MIT license. As of the collection time, it has 1881 stars on GitHub. This article covers its core architecture, typical use cases, and practical tips for getting started.

agent-sandbox: Manage isolated, stateful, singleton AI agent runtimes

agent-sandbox is an open-source project from Kubernetes SIG, designed to manage isolated, stateful, and singleton AI agent runtimes. Developed in Go, it offers declarative APIs and CRDs, simplifying agent deployment and operations. It is ideal for AI applications requiring long-running, persistent state, and has over 3100 stars on GitHub.