Recently updated open-source projects

Hush is a Safari content blocker for iOS and macOS that removes many cookie-consent and privacy-tracking overlays. After installing it, enable the blocker in Safari settings; its rules run locally without access to browsing habits, passwords, or crash reports. It does not click consent buttons, decide cookie permissions, or block ads. Some sites may still show notices when hiding them would break the page or when the site actively resists blocking.

fizzy-cli brings Fizzy boards, cards, comments, search, attachments, activity, and related administration into a terminal interface that shell-capable AI agents can also use. Commands return structured JSON with breadcrumbs that suggest useful follow-up actions, and the repository includes setup for a Claude plugin and embedded agent skill. Initial configuration requires a Fizzy personal access token and account selection, with an optional default board.

Collect single-choice, multi-choice, and free-form questions from parallel AI agents in a separate terminal inbox, then route your answers back to the correct sessions. AUQ integrates through MCP, an OpenCode plugin, or Agent Skills and works over SSH with native notifications; Bun is recommended for its default terminal renderer.

ProtonSearch is a local-first Windows launcher opened with Alt+Space. One search box can find installed apps, indexed files and document text, browser history, clipboard items, OCR from images, Git activity, notes, settings, commands, and agent chats. Its indexed database stays on the PC, while optional prefixes, calculator results, window controls, quicklinks, and reusable snippets support keyboard-first daily work.

ClawFu packages positioning, copywriting, persuasion, sales, SEO, customer research, revenue operations, and other marketing methodologies as agent-readable skills. Users can connect the npm MCP server to Claude, Copilot, Cursor, or Windsurf, or consume individual SKILL.md files through compatible tools. The collection is intended for teams that want agents to follow named methods, clear triggers, and defined output formats instead of relying on one-off prompts.

Harnessclaw is an Electron desktop application for configuring AI agents, chatting with them, browsing skills through ClawHub, and reviewing session history from a single interface. It is aimed at users who want a visual control surface instead of managing each agent workflow separately. The current documentation focuses on installing dependencies and building from source with Node.js and Yarn, with build targets for macOS and Windows.

Agent Skills Playground contains experimental skills and demos for MATLAB and Simulink work, including skill engineering, model-based systems workflows, and embedded AI deployment. Individual skills can be copied into Claude or installed according to another compatible agent's conventions. The repository is explicitly a sandbox: content may be incomplete, change without notice, or move elsewhere, while official use should favor the MATLAB and Simulink Agentic Toolkits.

This repository packages 20 modular Agent Skills for idiomatic Go development, distilled from Google, Effective Go, Uber, and community guidance. Coding agents can load procedures for code review, concurrency, context handling, errors, interfaces, naming, performance, testing, and related decisions, supported by 48 references and eight check scripts. The collection works across multiple coding tools through the common Agent Skills installation flow. Version-sensitive recommendations are tracked separately so users can check minimum Go releases and fallbacks.

Determine whether a Chrome, Edge, Chromium, or Electron environment can decode or encode HEVC in hardware before shipping video features. The guide maps codec profiles, HDR behavior, operating systems, GPUs, drivers, and browser versions, then shows capability checks with MediaCapabilities, Media Source, WebCodecs, WebRTC, and MediaRecorder. It also documents patches and build paths for custom Chromium or Electron distributions. Results vary substantially by platform and version, so browser branding alone is not a reliable compatibility test.

Jitsi Meet Electron is a cross-platform desktop client for connecting to any Jitsi Meet deployment. It adds automatic updates, screen sharing, an always-on-top window, and deep links that open a room on a configured server. End-to-end encryption is marked beta, while remote control is disabled because of security issues; installers are available for Windows, macOS, and Linux.

ffmpeg-commander turns common video and audio encoding choices into a browser form that produces an FFmpeg command. It helps users assemble routine operations without memorizing the tool’s large option set. The generator intentionally does not cover every FFmpeg capability and makes assumptions, so its output may need adjustment for a particular version or build configuration. An optional companion service, ffmpegd, can execute submitted jobs and return live progress to the browser.

Meru, formerly Gmail Desktop, puts Gmail in a dedicated cross-platform desktop window for users who do not want their inbox buried among browser tabs. The project provides downloads for macOS, Windows, and Linux and is built with Electron. Meru is a third-party client that is not affiliated with Google; Gmail remains the underlying account and mail service.

Produck Skills is an open-source bundle for coding agents that turns vague user requests into scoped, testable PRDs and carries real feedback into implementation work. The collection also includes workflows for producing animated product demos and improving website visual judgment through screenshot-based review. It can be installed as a bundle across Codex, Claude Code, Cursor, and other compatible agents. The feedback workflow needs a Produck MCP token to load ticket context, while the remaining skills can be used without it.

AI Hot Monitor is a hands-on teaching project built around a working trend-tracking application. Users define keywords, aggregate results from more than eight sources including Twitter, Bing, Hacker News, and Bilibili, then apply AI query expansion, authenticity checks, relevance scoring, and summaries. Results can be filtered and pushed through WebSockets or email, and the monitoring workflow is also packaged as an agent skill; the repository is primarily presented as a build-from-scratch course project rather than a hosted service.

OpenZeppelin Skills gives agents structured workflows for secure contract development using the OpenZeppelin Contracts ecosystem. The available skills cover project setup and contract upgrades across Solidity, Cairo, Stylus, and Stellar, while the secure-development workflow can use the OpenZeppelin Contracts CLI to generate reference implementations. Users can install the collection through the Skills CLI, a Claude Code plugin, or manual copying, making it most relevant to teams already building with OpenZeppelin libraries.

ishare is a macOS capture utility for taking screenshots of a selected region, a window, or an entire display, as well as recording video or GIF output. Recording codecs and compression can be adjusted before sharing. Its distinguishing workflow is configurable uploading: an .iscu JSON file can describe a custom server or service, request headers and form data, the returned media URL, and an optional deletion link. Confirmed targets include self-hosted file services and Discord webhooks.

Install this collection to give compatible coding agents reusable guidance for Kotlin projects. Each skill packages its own instructions, scripts, and resources, and can be added through the Skills CLI, Claude Code or Codex marketplaces, or copied into a project manually. The current catalog focuses mainly on backend development and tooling rather than providing a standalone application.

Share a precise part of your Windows screen even when a meeting app only supports whole-window sharing. The tool opens a transparent, click-through capture frame and a separate rendering window; resize or move the frame, then select the rendering window in Teams or another conferencing app. A Windows executable is available from releases, while source builds and command-line options let users preset coordinates, dimensions, display scaling behavior, and maximum capture frame rate.