Recently updated open-source projects

allincms-content-ops organizes the path from raw sources and a living wiki to search intent, web drafts, media, review, publishing, and indexing. Humans and supported coding agents can share the same file-based workflow, while release gates require traceable sources, complete publishing fields, audits, and an independent adversarial review. Full execution is best suited to environments with browser automation and subagent support.

Kindle Cover Skill takes a front-cover image, title, author, page count, and optional back-cover copy, then calculates the spine and builds one KDP-ready PDF containing back, spine, and front. It supports several trim sizes, white or cream paper, configurable image fitting, bleed and safe zones, and requires ReportLab plus Pillow; Amazon adds the barcode at print time.

genai-prompt-optimizer finds every LLM call in a GenAI app, pins one model per run, and derives adversarial cases from the app’s prompts and JSON schemas. It first records the expensive model as an oracle, then iterates on the cheaper student’s prompt and re-verifies each cohort with deterministic checks, semantic comparison, and limited judge review. The result reports measured token cost and projected savings; it asks for consent before the token-heavy run.

visualizing-data helps an agent choose a chart from a message-first framework and a catalog of more than 24 chart types. It covers decision trees, annotations, operational dashboard patterns, perception principles, colorblind-safe color systems, performance, and HTML or document output. The emphasis is on making the intended message legible and accessible, not merely decorating a dataset.

hekouwang-channel-analyst reuses your Chrome login session to pull your own Xiaohongshu creator-backend data into CSV time series. It analyzes the view-to-profile-to-follow funnel, out-of-pool diagnosis, note rankings, trends, and follower-target levers, then builds a self-hosted ECharts HTML report with one priority action and one explicit “do not do.” Historical recommendations are compared with later results; it does not inflate metrics or promise follower numbers.

trip-planner builds an itinerary around geography, dates, and logistics before filling it with attractions. It red-teams the plan for long transfers, closures, holiday traffic, weather, vehicle capacity, and other failure modes, then verifies places against the actual travel dates. The output is a clean group-friendly PDF with route context, book-now warnings, and Google Maps links that open when a traveler taps each stop.

humanizer-it looks for Italian-specific problems such as English syntactic calques, nominal bureaucratic style, missing discourse markers, and weak use of ci or ne clitics. It supports full editing, diagnosis-only audits, and targeted fixes. A dependency-free Python scanner counts banned constructions, sentence rhythm, and Italian morphosyntactic signals, then reports a cleanliness score.

Upload PDF, spreadsheet, or screenshot reports from Russian brokers to consolidate positions across accounts. The skill uses current MOEX, Bank of Russia, and cbonds data to assess duration, after-tax yield, concentration, and possible replacement candidates, with optional Monte Carlo risk scenarios. It focuses on Russian bonds and equities, does not execute trades, and presents analysis rather than personalized investment advice.

Designed for teams packaging ERP, CRM, HIS, and similar workflows, the tool probes each step for data sources, decision logic, and failure handling before generating the skill. External interfaces are wrapped in Python scripts with mock support, followed by evaluation and iteration. It also supports editing existing skills, changing deployment parameters, and importing reusable domain knowledge.

paper-maestro is a Claude Code orchestrator for moving an academic manuscript from idea and literature gap through structure, drafting, data, citation verification, review, and submission. It is tuned to Russian VAK and GOST requirements, with a separate English or Q1 track, multi-agent barriers, and quality gates for overclaiming, reproducibility, and unsupported citations.

codex-image-gen drives an interactive Codex pane through tmux, requires an explicit absolute save path, waits for the file, and verifies the image and target aspect ratio with ffprobe or PIL. If the expected file is missing, its helper searches the Codex generated-images cache and can recover base64 data from the newest session rollout JSONL. It is a reliability wrapper around an interactive workflow, not a standalone image model.

dreamifly-batch turns a prompt list into repeatable Dreamifly image or video jobs. Before spending credits, it validates parameters, estimates cost, and offers a dry run; after generation it saves media, metadata, and structured results locally with caching, retries, and resumable failures. Expensive video or premium-model jobs require confirmation.

SSW.YouTubeWatchStats uses a dedicated persistent browser profile to read roughly 30 days of YouTube history and write a developer-focused Markdown report. It separates technical from non-technical viewing, groups topics and formats, and estimates watch time and leading channels. History stays on the machine, but the skill requires an interactive signed-in browser window, so it will not run in a headless environment without a display.

skill-planok packages the PlanOK CRM REST API documentation in Spanish and English for Claude Code and other Agent Skills runtimes. It covers endpoints, authentication, parameters, schemas, and examples across the CRM modules, and includes a local CLI for keyword searches. The skill explicitly refuses to invent an endpoint or field when the documentation does not declare it.

chalktalk teaches an AI coding agent to turn a math topic into an end-to-end animated explainer with scene design, rendering, and frame checks. It uses Pango text and Unicode for equations, avoiding a large TeX installation and its common rendering traps. Setup scripts prepare the Python environment, animation engine, Cairo, Pango, and FFmpeg on macOS or Linux; the result is a rendered video rather than an editable lesson deck.

Bootstrap an Angular 21 and Tauri 2 desktop application without wiring the frontend and native shell together from scratch. The template provides TypeScript, Sass, hot reload, a browser-only development mode, and commands for producing operating-system-specific bundles for Linux, Windows, and macOS. Its simple project structure separates the Angular renderer from the Rust-based Tauri process. Development requires Node.js 22 or later, a Rust toolchain, and the platform prerequisites listed by Tauri before the sample can run or package successfully.

level is a convenience package that exposes a common abstract-level interface while selecting classic-level in Node.js and browser-level in web applications. It supports lexicographically ordered keys, batch writes, asynchronous iteration, encodings, and TypeScript declarations. Browser projects normally use it through a bundler such as Browserify, Webpack, or Rollup, while Node.js stores the database in a local directory.

Go Lean SDD sizes planning to the change: minor work stays in conversation, while larger features earn durable artifacts. Every change includes colocated tests plus a mutation-style discrimination check that proves the test can fail, and the agent searches the project, dependencies, standard library, framework, and platform before adding new code. It keeps durable context in one AGENTS.md instead of a stale scaffolding tree.