ritmex-skills

Detect, review, prevent, and remove AI-generated UI convergence patterns ("AI slop") in frontend code. Use whenever building or styling a UI, reviewing frontend changes, refining visual design, or when the user asks to remove the AI look, make an interface feel human-designed, polish a page, audit UI quality, or explain why a design feels generic. Covers HTML, CSS, JSX/TSX, Vue, Svelte, and Tailwind; audits 46 catalogued patterns plus composition-level checks, separates source-confirmed, rendered, and judgment-based evidence, and rewrites authorized code toward a product-specific design direction without requiring external dependencies.

Teenage Engineering / Dieter Rams functionalist-hardware aesthetic: neutral molded panels, a single bold accent color, tactile knobs and buttons, LED and segmented displays, uppercase monospace labels, visible "device chrome" (screws, bezels, serial numbers, power LEDs), and playful retro-futurist technical detailing. Triggers: "TE style", "OP-1 style", "Pocket Operator look", "Braun / Dieter Rams style", "functionalist hardware UI", "retro hardware / synth / audio-gear interface", "device-like UI", "tactile / physical interface", "minimal skeuomorphic panels", or descriptions involving cream/charcoal panels, knobs, screws, LED dots, a single accent, monospace labels, phosphor/LCD screens, or anything that should look like hardware.

Auto-trigger this skill when the user's input is vague, ambiguous, incomplete, or too brief to act on reliably in Codex. Indicators: single-sentence requests with no success criteria, missing target files or scope, unclear intent ("fix this", "make it better", "help me"), mixed unrelated asks in one message, or prompts that would force Codex to guess critical details. Do NOT trigger when the user's intent, scope, and done-state are already clear — even if the prompt is short.

去除文本中的 AI 生成痕迹。适用于编辑或审阅文本，使其听起来更自然、更像人类书写。 基于维基百科的"AI 写作特征"综合指南。检测并修复以下模式：夸大的象征意义、 宣传性语言、以 -ing 结尾的肤浅分析、模糊的归因、破折号过度使用、三段式法则、 AI 词汇、否定式排比、过多的连接性短语。

React best practices for component design, state management, and Effect discipline. Use when writing, reviewing, or refactoring React components, custom hooks, or any .ts/.tsx/.js/.jsx file that uses React.

Apply when the task requires implementing net-new behavior — features, modules, endpoints, integrations, or any additive change. Drives development through strict red-green-refactor with test-first slicing and deterministic verification.

Spawn a subagent to produce high-quality project documentation (README, integration guide, tutorials, product docs) from the current repo state.

Improve UI/UX aesthetics and hierarchy (color, spacing, typography, layout, components) from UI code or specs. Trigger when the user asks to beautify/polish/refine/modernize the UI, wants a minimal/enterprise look, or reports the UI is messy/inconsistent/cluttered.

Track recently updated airdrop projects from CryptoRank (last 3 days). Save report locally, then send a concise Telegram notification.

Generate or update a .gitignore file based on the current project stack and tools. Use when the user invokes /ignore or asks to create, fix, or audit a .gitignore for the current repository.

Design and generate production-quality SVG logos in modern minimalist style. Use this skill PROACTIVELY whenever the user asks to create a logo, design a brand mark, generate an SVG icon, make a logomark, create a wordmark, build a brand identity symbol, or needs any kind of vector logo. Also triggers for requests like "make me a logo", "design an icon for my app", "create a brand symbol", "I need a logo for...", "generate SVG logo", or any task involving logo/icon/brand mark creation — even if they don't specifically mention "SVG" or "minimalist".

Control Slack via the `slack` CLI to read, search, and manage messages, threads, files, reactions, channels, DMs, and canvases. Trigger on requests involving Slack messages, threads, URLs, channel history, unread or recent DMs, or sending/replying to messages (English or Chinese queries mentioning Slack).

Use for ClickUp tasks and docs. Trigger on ClickUp URLs, task IDs, list IDs, doc IDs, page IDs, assignees, statuses, due dates, comments, checklists, and document workflows. Use the global `clickup` CLI.

Apply this skill whenever the user asks to debug, review, audit, or find problems in code — regardless of language or framework. Triggers include: "find bugs in this", "review my recent changes", "something is wrong but I don't know where", "check for security issues", "audit my commits", "why is this slow", "what am I missing in my tests", or any request to investigate correctness, safety, or performance. Prioritize recently modified files, staged changes, and the latest commits — the bug is almost always in what just changed.

Apply this skill whenever the user asks to refactor, clean up, restructure, or improve existing code — regardless of language or framework. Triggers include: "refactor this", "clean up my code", "this file is too long", "this function is hard to read", "reduce duplication", "decouple this", "improve maintainability", "apply design patterns", or any request to improve code quality without changing observable behavior. Also trigger when reviewing code that has obvious smells: god files, deeply nested logic, long parameter lists, duplicated blocks, or functions exceeding ~30 lines.

Use when the user asks about OpenAI Codex CLI (codex, codex exec, codex resume, codex review, codex cloud), wants to run code analysis, refactoring, or automated editing via Codex, needs help configuring AGENTS.md, config.toml, MCP servers, Skills, or Automations, or asks about Codex best practices, approval modes, sandboxing, or multi-agent workflows.

Fetch up-to-date library documentation via the ctx7 CLI. Use PROACTIVELY whenever any code change, feature design, or implementation or user request involves a project dependency — always query the matching version's docs first before writing code.