Recently updated open-source projects

Shinri designs prompts for expressive calligraphy and lettering images with Nano Banana 2, Nano Banana Pro, or ChatGPT Image2. An interactive flow starts from Chinese or Japanese characters, longer text, or a signature, then combines eight style families with five emotional axes covering weight, texture, temperature, spatial fullness, and time. It produces prompts rather than images and is intentionally not aimed at standard decorative lettering.

Complete a guided self-discovery interview, which can be split across sessions, and the skill produces two Markdown files: one describing your context and patterns, and another defining how you want an AI to collaborate with you. Paste them into future conversations to reduce generic advice and re-establish context across tools. The interview allows skipped questions and is designed not to collect direct identifiers such as your real name, phone number, ID, exact birthdate, or home address.

Ebou groups new Mastodon posts by author and presents them in a messenger-like interface, making friends’ updates easier to scan while keeping replies in a nested conversation tree. It also covers timelines, posting with image or video attachments, boosts, favorites, bookmarks, lists, profiles, and basic search. The client remains experimental and incomplete: macOS is the stable target, Windows is in beta, and Linux is theoretically supported but untested. Pleroma compatibility is also described as untested.

This static Agent Skill helps an assistant advise you on recording a voice performance for conversion into instruments such as piano, guitar, violin, drums, or saxophone. It covers microphone and environment choices, instrument-specific technique, prompt writing, file formats, and diagnosis of muddy, robotic, off-rhythm, or clipped output. The skill supplies guidance only; it contains no executable conversion code, telemetry, or backend calls.

ziyu packages crisis-communication playbooks for brands, creators, and small teams using Claude Code. It covers eleven crisis types with severity assessment, information checklists, response strategies, action lists, recovery timelines, and templates for apologies, clarifications, denials, and legal warnings. Separate monitoring guidance supports current-issue scans and scheduled checks that organize signals into a structured public-opinion report.

trade-chokepoint evaluates import and export lanes for supply-chain bottlenecks, landed cost, and trade risk. Its India edition incorporates factors such as CEPA, RoDTEP, and BIS, applies weighted scores and penalties, and produces a one-page PDF verdict. The project presents the result as research support: its weights are not empirically validated, and the output does not replace customs, legal, or tax advice.

Use one skill to assemble quantitative Polymarket bots with Kelly sizing, Bayesian updates, liquidity checks, and multi-source signals, and another to score wallets and flag whale, clustering, or possible insider activity. Templates, scanners, backtesting, and terminal alerts are included. The project is educational and recommends dry-run testing before risking capital.

Source a session transcript, file, or topic, extract the concepts that actually matter, and turn them into a dated checklist. The skill asks one question at a time, probes why a decision was made, marks confirmed items immediately, and rephrases questions when needed. It does not offer to finish until the checklist is complete, with an optional mode for teaching another person.

Consulting Visual helps create strategy-consulting covers, executive infographics, methodology diagrams, flowcharts, and 2×2 analysis pages. It separates the logic of covers from information and analysis layouts, condenses long themes into a title system, and chooses a clear business metaphor such as a path, funnel, flywheel, or matrix. Code-rendered HTML, CSS, and SVG keep the typography sharp across several aspect ratios.

kens-designer steers Claude toward senior design-engineering practice when building or critiquing interfaces, with React and Tailwind as the default stack. It begins by learning the desired visual direction, then combines implementation with a critique of hierarchy, layout, typography, color, accessibility, and interaction details. The repository also describes an AGENTS.md setup for Codex and distills principles from a broad design reference library.

This Claude Code skill fetches public WeChat articles from the user's machine, searches by topic, reads selected results, and caches repeated requests locally. It is designed for public `mp.weixin.qq.com/s/...` share URLs and does not cover articles restricted to followers or other non-public pages.

Turn one-line LaTeX formulas into clean, printable images that can be pasted into Word, WPS, or Google Docs. The tool runs either as a Claude Code skill or a standalone local page with live preview, PNG clipboard copying, and SVG copy or download. It needs only Python 3 and a modern browser, with bundled MathJax for offline use and no LaTeX installation.

Give the skill a company name and it searches review sites, app stores, Reddit, videos, and the wider web for recurring product friction and usable customer quotes. It packages the findings into improvement ideas and pointed questions for design interviews, competitive analysis, or product critiques. It is not useful for pre-launch products that lack a meaningful public review surface.

Connect to IB Gateway to inspect quotes, option chains, positions, portfolio Greeks, P&L, concentration, earnings exposure, and wheel cycles, or preview how a proposed trade would change risk. Most scripts are read-only; order placement is isolated behind two explicit opt-in gates. Real-time Greeks require the relevant IBKR market-data subscriptions, and the output is intended for research rather than financial advice.

This Claude Code skill uses BBDown to save a single Bilibili video or batch-download an uploader's videos as high-bitrate M4A files. It can embed available subtitles, names files with the date, title, and BV identifier, and skips completed downloads for resuming. BBDown must be installed, and some content requires a valid logged-in cookie.

autoresearch-skill gives Claude a disciplined loop for optimizing a measurable target such as test runtime, bundle size, latency, memory, or training loss. Each experiment is benchmarked, useful changes are committed on one branch, and regressions or failures are reverted while the run history remains available. Session files support resuming later, but the approach is intended for iterative work rather than trivial one-off edits.

Circuit reads Gmail in a read-only flow, keeps threads with genuine two-way replies, filters newsletters and automated senders, and surfaces relationship type, status, deadlines, and action items. The open-source version can use Claude's Gmail connector or a local MCP engine, with no Circuit server handling the mail. Background sync and a web dashboard belong to the hosted tier.

spring-boot-pr-review-skill reviews Java and Spring Boot changes for production failures that generic reviewers often miss. It looks for bypassed proxy annotations, silent transactional-event behavior, N+1 queries, tenant-unsafe cache keys, missing optimistic locking, and external calls held inside database transactions. Findings explain the underlying mechanism, assign a calibrated severity, and suggest a concrete fix rather than focusing on superficial style.