Recently updated open-source projects

Research a company by opening its site, verifying funding with at least two independent sources, checking customer overlap, and separating technology from packaging, sales, brand, and network effects. Recommendations receive a six-dimension score, while fundraising references stay separate from practical learning targets. Historical judgments are marked when they were not re-verified.

BudgetBudget is a macOS desktop app for planning monthly budgets alongside transactions from MoneyMoney. Install it from the provided DMG, follow the in-app setup, and use the budget view to make financial decisions with more context. It is aimed specifically at MoneyMoney users, and the maintainer notes that the project is developed in spare time, so issue and pull-request responses may be slow.

Turn a rough prompt into a ready-to-paste version tailored to the task, intended model, constraints, audience, and output format. The skill selects only techniques that add value and adjusts its rules for Claude, OpenAI reasoning models, and Gemini rather than applying one generic template. Each delivery includes a concise changelog explaining the fixes, and the skill improves the prompt without running the underlying task unless explicitly requested.

TikZ Scientific Figures turns a plain-language description into editable TikZ or PGFPlots source, compiles it into a vector PDF, and returns both files. It covers plots, circuits, biological and physical diagrams, and other technical figures, with optional SVG editing through bundled tools. A LaTeX toolchain such as pdflatex is required, with extra dependencies for previews and SVG export.

Maestro runs a self-contained pre-release review across scope, root cause, security, logic, code quality, frontend, database, Python, performance, memory, hygiene, and verification. It orders the applicable checks, applies the rule of investigating root causes before fixes, and combines the results into one executive ship report. Full, Quick, and Incident modes let teams choose the breadth of the review.

Draft Korean certified-mail notices for cases such as unpaid wages, deposit recovery, contract termination, and damages, with 14 supported categories and a court-oriented DOCX layout. Its MCP flow masks names, addresses, and amounts before Claude sees them and restores them locally into TXT and DOCX. Without MCP, the project recommends placeholders, and the result remains a draft to review.

clarity-first applies a First-Time User Experience lens to every screen and asks whether each element is immediately understandable, needed now, excessive, and reducible. It returns concrete keep, cut, redesign, or fix decisions instead of vague UX advice, making hidden jargon and onboarding friction easier to spot.

Holo-Zlib is a portable Agent Skill for searching and downloading ebooks through the Z-Library eAPI, from Claude Code, Codex, Cursor, Gemini CLI, or a shell. Runtime use needs only Python 3.10+ and the standard library. Credentials and downloads stay outside the skill directory, but users must prepare credentials/zlib.json and can choose an output directory.

Vastu-Jyotish Mentor uses birth and business details to calculate charts, planetary periods, auspicious times, Vastu considerations, and numerology, then returns a short verdict with plain-language reasons or suggested remedies. It uses pyephem and keeps personal data in a local private profile. The project frames this as a traditional cultural lens, not a scientific guarantee or a basis for medical, investment, or marriage decisions.

Humanamente rewrites Spanish AI text by identifying inflated phrasing, negative parallelism, mechanical lists, English punctuation, generic conclusions, and other local writing tells. It preserves meaning while iterating through a six-dimension style check, without regex, scripts, or find-and-replace rules. A simple mode is available, with optional voice profiles for close, corporate, technical, or commercial tones.

The Council is a Claude Code panel for validating one idea or comparing two options. Permanent critics and cast domain specialists research in parallel, while a Chair produces a structured verdict with objections, blocking conditions, and next actions. Runs offer Quick, Standard, and Deep effort levels, resumable sessions, a logbook, and Markdown or HTML reports; extra critics need their own CLIs or APIs.

Chinese EPUB Typo Fixer handles OCR mistakes in Chinese ebooks made from scanned images. It applies clear visual-character or simplified/traditional corrections by default, lists archaic-usage modernization separately, and outputs a corrected EPUB plus a change report. The scripts only unpack, rewrite, and repack; they avoid network access and skip ambiguous or oversized changes.

Write AI News turns a supplied URL or news title into Thai AI-news copy: a headline of up to 60 characters, a one-sentence summary, three to four body paragraphs, tags, and source information. It is a formatting and drafting skill for Claude Code; the project description does not promise automatic live-news retrieval or verification.

Moto News is a Claude Code plugin for drafting content about Harley-Davidson and Indian Motorcycle. It covers new-model news, motorcycle reviews, event coverage, riding culture, and social posts for channels such as Facebook, Instagram, and X. The repository describes content-generation prompts, not live news retrieval or automatic publishing.

kuangjia-chaijie helps developers understand a cross-module framework topic before writing a talk or article. It combines top-down mapping with bottom-up symbol tracing, scope checkpoints, phase exit criteria, and clean-context cross-checks, then produces structured Markdown notes rather than the article itself. It is not intended for single-function debugging.

Describe a task in plain language and Automation Level Advisor interviews you to decide whether AI should run alone, be supervised, require approval at each step, or remain an assistant to a human-led process. Built for non-developers, it explains the reasoning and creates matching Markdown and HTML project-starter reports with checkpoints, guardrails, phased rollout guidance, risks, sources, and example prompts for Claude Code or OpenAI Codex.

Use anneal when several plausible approaches compete and intuition may mislead. It turns the goal into measurable questions, sketches two or three directions, prototypes them cheaply, selects by a direction-fitness score, and only then applies a separate polish rubric. It is a lightweight human decision discipline, not an autonomous engine, and this repository is a legacy mirror.

compass audits a long Claude Code session along two axes: drift from the original task and rot in the codebase. It reads the transcript plus cached Git, test, lint, file-size, TODO, and boundary signals, then returns a SAFE, CONSIDER, or STOP recommendation with a next action. It does not run tests or research the web, supports Claude Code JSONL transcripts only, and is now a legacy mirror.