Recently updated open-source projects

Describe a time range, scene, or output goal in Claude Code to cut footage, remove silence, turn long videos into Shorts, or generate and style subtitles. Changes are previewed through a dry run and backed by project backups and rollback guidance. The skill depends on the unofficial capcut-cli and CapCut or JianYing Desktop, so CapCut Web is unsupported and application schema changes may break compatibility.

Novel Creation Omnibus is a modular writing workshop for long web fiction, short stories, dialogue, character arcs, world systems, pacing, platform adaptation, and revision. You can load only the tools for a task, from outlining and opening hooks through chapter drafting, continuity tracking, and multi-layer editing. It includes genre and model-specific approaches, but does not silently rewrite or delete a project’s outline, character files, or other source material.

shopping-aggregator triages a purchase by product, region, budget, urgency, and condition, then delegates live research across thirteen consumer-shopping domains. It ranks landed cost rather than sticker price, pins the exact variant and timestamp, checks stock and coupon claims in the cart, and surfaces seller trust tiers, coverage gaps, and disagreements between snapshots. It is an orchestration layer rather than a price-fetching engine, so missing research tools become explicit gaps. The skill recommends what to buy but never completes the purchase.

Manim Skills guides an agent in creating math, physics, and computer-science animations with Manim Community Edition. It covers scenes and mobjects, MathTex equation transformations, plotted functions, ValueTracker updates, geometry, 3D surfaces, and moving cameras, then renders to MP4 or GIF. A render-a-frame-first check helps catch layout problems before a full export. The workflow requires a Manim and LaTeX-capable Python environment.

Loop Engineering turns an ambiguous coding request into five stages: frame, model, gate, build, and verify. Each slice starts with a check that should fail, moves to green through small changes, and ends with tests, type checks, builds, and an independent verifier. Frontend or deployment work also needs a real runtime probe, while high-risk slices cannot skip independent verification.

Start with confirmed bookings, scattered notes, or no reservations at all. The workflow compares route and pacing options, confirms the itinerary structure, then verifies time-sensitive details such as opening hours, fares, events, and transport before assembling a mobile-first single-file HTML handbook and printable version. Unbooked items remain clearly labeled, with a prioritized list showing what to reserve first.

promotion-assistant coordinates email, posts, forum replies, and direct messages for a shipped product, tracks a six-layer conversion funnel, and uses Thompson Sampling to adjust tactics from measured feedback. It keeps copy, audiences, policies, and credentials in a separate private config repository. Dry-run is the safe default: no outbound action is allowed until the product and that individual channel are explicitly enabled, so it is neither a spam cannon nor a scheduling engine.

This skill routes Claude toward terse, high-signal responses with lite, dev, and ultra levels. It keeps more explanation in lite, compresses routine development work in dev, and removes most framing in ultra, while learner-facing and training content exits compact mode. It changes response style, not task capability.

Use this Claude Code skill to plan and carry out an eight-stage paper workflow, including topic refinement, drafting, polishing, first-pass review, defense preparation, and submission. It handles general writing work directly, then recommends and hands off to specialized skills for tasks such as deep literature research, framework figures, or institution-specific document formatting. Its built-in checks emphasize source-grounded claims, hypothesis discipline, and readable prose.

Provide a visual mood and music direction, and the skill finds a track, detects strong beats, plans short shots, generates frames with Seedream, animates them with Kling, and assembles the result with ffmpeg. It is designed for roughly 20–30 second, 3:4 atmospheric vertical videos rather than dialogue-heavy or performance-driven scenes. The current workflow requires macOS and paid Seedream and Kling API access.

make-playbook helps an indie maker diagnose where a product is stuck across idea, build, launch, growth, monetization, automation, and exit. Claude Code loads the relevant checklist, asks questions about the actual app, and ends with a concrete next step; an optional MAKE.md file tracks progress over time. It is an original companion inspired by Pieter Levels' MAKE, not a substitute for the book.

This skill draws visually rich slide bodies in HTML and inline SVG, renders them to PNG with headless Chrome, and embeds the result through python-pptx. Headlines, footers, and page numbers remain editable native text while shadows, gradients, curves, arrows, and icons are rasterized consistently. The workflow includes verification through Keynote or LibreOffice and builds on Anthropic's PPTX skill.

escrita-natural helps Claude rewrite Brazilian Portuguese from meaning rather than word-for-word English structure. It targets translationese, false cognates, stiff syntax, inflated chatbot phrasing, and generic enthusiasm, then applies tone calibration and a final human-sounding pass; it is aimed at prose rather than code, raw data, or strictly literal technical text.

HWPX Skill helps an agent draft Korean administrative documents and output them as HWPX files, following the Ministry of the Interior and Safety’s AI-era writing guidelines. It combines Korean public-sector document drafting with file generation in one workflow for Claude Code users.

X Copywriter is a multilingual skill for X content, account appeals, and pre-publication compliance checks. It drafts several hooks, tests the single point and intended action, then revises weak dimensions before delivery; a separate humanizer reduces generic AI phrasing. Appeal writing is constrained to verifiable facts and rejects invented excuses, making the workflow more disciplined than a generic copy prompt.

This repository is a reference implementation of the RealWorld application using Rust, Yew, and WebAssembly. It demonstrates a full frontend with CRUD flows, authentication, routing, pagination, function components, and hooks while following the shared RealWorld API specification. Developers can run the web build through Docker or the Rust and Trunk toolchain, and optionally package the same application as a Tauri desktop app. It is intended for learning and comparison rather than as an end-user service.

Lovcode gives developers a desktop workbench for launching Claude Code, Codex, terminal, and other AI coding sessions from a selected project. It indexes local histories and exposes tool calls, linked files, generated images, and usage details, making previous work easier to find and resume. Users can copy handoff context between agents and manage commands, MCP servers, skills, hooks, runtimes, and model providers from one interface. Desktop releases target macOS, Windows, and Linux.

GB/T 9704 Format standardizes applicable government-document layout rules in Word files through a Claude skill or standalone command-line tool. It fixes margins, headings, fonts, spacing, page numbers, punctuation, numbering, and table formatting while leaving the document text untouched. Uncertain content issues are reported rather than guessed, and the Python-based tool does not require Word or WPS.