Recently updated open-source projects

This Claude Code skill examines a digital project under Russian law and returns three views: technical and security risks, a lawyer-ready document brief, and business readiness. It derives processes from existing code or asks questions for a new idea, then highlights priorities and launch blockers without presenting legal advice.

GitLight combines GitHub and GitLab notifications in a single kanban-style inbox instead of making developers check each platform separately. It can deliver push notifications, show detailed event information, and filter activity by repository, organization, pull request, issue, commit, and other project context. Both services can be monitored at the same time, which is useful for teams whose work spans multiple hosting providers. The free open-source client is available for macOS, Windows, Linux, and the browser, with a Homebrew installation option on macOS.

Use pixeltamer from Claude Code, Codex CLI, OpenCode, Cursor, or another skill-aware agent to generate, edit, or compose images with gpt-image-2. It structures prompts, selects an available backend, and reloads each result for a visual check before reporting success. Authentication can use an OpenAI API key or a signed-in Codex client. Mask-based inpainting and parallel variant generation require the API backend, while standard editing and composition can use either backend.

AutoCodeRover tackles GitHub bug reports and feature requests by using program-structure-aware search to locate relevant files, classes, and methods before generating a patch. When a test suite is available, it can use fault localization and regression testing to improve patch selection. It works on local projects and also includes a SWE-bench experiment mode, with a Conda environment and a supported model API key required.

Design Review evaluates a page by looking at rendered screenshots rather than guessing from source code. Through Playwright MCP it captures desktop, tablet, and mobile views, checks typography, contrast, spacing, hierarchy, color, alignment, and responsive behavior, then proposes ready-to-apply CSS fixes. It can rerun the audit after changes, but requires a connected Playwright MCP server.

skill-telegram uses Telethon and the Telegram user API to read Saved Messages, direct chats, groups, and public channels as JSON lines. It can batch channel posts by date, sanitize text, and download voice or audio files for later processing. The skill is read-only, requires an interactive first login, and stores a sensitive session file; downloaded voice notes are not transcribed automatically.

ai-video takes a request from first-frame generation and prompt writing through rendering, clip review, voiceover, assembly, and upscaling. It routes a shared request format to six supported video models and records lessons from each quality-control pass. You need API keys only for the services you call, plus ffmpeg and ffprobe locally.

council-diff runs five differently biased personas in parallel on the same decision and exposes their scores, verdicts, strengths, gaps, agreement, and recommendation. Built-in domains cover founder, engineering, investing, career, product, and quant questions, while custom rosters are supported. An optional Oracle reviews difficult or split calls, and the Brier module scores predictions after outcomes resolve. Safe mode can keep the Oracle on a zero-retention model when retention matters.

Prompt Mentor turns a rough request into a clearer instruction before carrying it out. It activates only when invoked, scales its explanation from a quick fix to a multi-file project, and shows the revised prompt, implementation, concepts, takeaways, and final prompt used.

E-Commerce Video Skills turns product or app screenshots into animated demos with device frames, cursor actions, zooms, and feature captions. It also creates short sale spots from prices and discount data, or turns a photo folder into a music-synced slideshow with Ken Burns motion and captions. Templates can be rendered across many products or offers, making the pack useful for ecommerce teams that already have screenshots, prices, and image assets.

LoFloccus is a Windows and macOS companion for the Floccus browser extension. It exposes one chosen folder through a small, self-contained WebDAV service, allowing bookmark files to be synchronized further with Dropbox, Syncthing, rsync, or another tool that works with local files. Each setup receives a random port, username, and password, while access is restricted to one directory and a narrow set of bookmark-related file types. It is useful when browser restrictions prevent Floccus from writing directly to disk.

Add reusable, versioned slash-command skills to Claude Code for work such as code review, security audits, architecture design, testing, documentation maintenance, and multi-agent development. The marketplace supports selective enablement, automatic updates, and repository scripts for scaffolding and validating contributed skills. It requires an installed and authenticated Claude Code environment plus Git access to the marketplace source.

Amazon Q Developer CLI brings an agentic chat workflow to the command line, letting developers use natural-language requests while building applications. The open-source project is no longer actively maintained and will receive only critical security fixes; current feature development has moved to the closed-source Kiro CLI.

MCP-Zero is the open-source code accompanying a paper on active tool discovery and retrieval for autonomous LLM agents. It includes instructions for an MCP-tools dataset covering 308 servers and 2,797 tools, plus code under MCP-tools/build_data for constructing a custom server dataset. The data file must be obtained separately and placed at the path expected by the evaluation code.

AgentFigureGallery gives coding agents a real visual reference library instead of relying on prompt-only descriptions of scientific style. Search the browser gallery, mark examples as liked, rejected, or selected, and export those choices as a bundle for Claude Code, Codex, Cursor, or another agent to use while writing plotting code. An optional full-public pack contains 16,341 candidates across ten plot types, while a small built-in pack supports quick smoke tests. Included example workflows can export PNG, PDF, SVG, and combined previews.

yoonmoon separates Korean writing support into humanization, translationese cleanup, proofreading, restyling, and AI-writing detection, with an end-to-end route when several steps are needed. It preserves names, numbers, quotations, dates, and claims, and warns that detection is probabilistic rather than proof of authorship.

Claude Plugins combines a searchable registry with separate CLIs for Claude Code plugins and cross-client agent skills. It resolves public GitHub sources, installs skills globally or per project, and supports clients such as Claude, Cursor, Codex, Gemini CLI, OpenCode, and others. Claude plugins require Claude Code 2.0.12 or later, while skill support depends on the target client.

sci-figure renders molecules, crystals, slabs, and nanoparticles from common structure data, then composes them with native draw.io elements. Its Python modules cover atom rendering, reusable figure layout, low-level XML building, and export to PNG, SVG, or PDF while keeping the editable source beside each image.