Explorar skills

Scan written drafts for AI writing tells and rewrite them out without flattening voice. Detects cliched vocabulary, antithesis tics, dramatic colons, em-dash rhythms, hedging, sycophancy, stat-flavored vagueness, stock phrasing, and chirpy closers. Triggers when a draft is reviewed, edited, refined, polished, or finalized; when the user pastes content for feedback or a "thoughts?"-style review; wh

Cognitive science-based AI tutor that won't let you fake understanding. Implements a full teaching loop (Explain→Example→Check→Evaluate→Practice), concept-level mastery tracking with FSRS spaced repetition, learning profile diagnostics, and zero-hint quizzes. Reads source material (PDFs, docs, code, URLs) to teach from specific content, or teaches from AI knowledge. Use this skill when the user sa

Use when checking if an app violates Korean professional licensing laws before shipping, when a user asks about 자격사법 점검, 법률 리스크, 출시 전 법적 검토, 바이브코딩 법적 리스크, or mentions 변호사법, 의료법, 세무사법, 공인중개사법, 공인노무사법 violations. Legal check for Korean vibe coders.

Generates a structured Context Block that carries your full project state — stack, decisions, constraints, what's done, what's next — into any AI tool instantly. Use this skill whenever the user switches AI tools mid-project, starts a new chat and wants to continue work, says "I need to explain my project to another AI", mentions context loss, re-explaining, or starting fresh. Also triggers on phr

Encode and decode Motorola FLEX paging messages in audio WAV format. FLEX is the high-capacity successor to POCSAG, used by commercial paging carriers worldwide and still active on VHF/UHF networks. Implements 1600 bps 2-FSK with the real FLEX BS1 preamble, A1/A2 sync, Frame Information Word, BCH(31,21) + parity codewords, capcodes, vector words, and numeric/alphanumeric/tone message types. Suppor

Search scientific papers on biology, medicine, health, weight loss, psychology, nutrition, neuroscience, fitness, aging, sleep, gut microbiome, mental health, psychiatry, pharmacology. Use when user asks health/medical/biology questions, wants scientific evidence for a claim, or provides a PMID/DOI/bioRxiv link. Also triggers on "clinical trials", "испытания", "фаза 3". Triggers on questions like

Use this skill when working with Google Cloud CLI (gcloud), Google Cloud APIs, Google Search Console API, Google Workspace APIs, OAuth authentication with specific scopes, enabling GCP APIs, managing GCP projects, or making authenticated REST API calls to Google services.

Analyze Claude Code session history to find repeated workflows that should become skills, plugins, agents, or CLAUDE.md entries. Use when asked to find automation opportunities, analyze usage patterns, discover repeated workflows, or figure out what to turn into skills. Differs from claude-automation-recommender which analyzes codebases — this analyzes actual user behavior across all sessions.

Academic paper reviewing skill for ML/AI conferences. Use this skill whenever the user asks to review, critique, or write referee reports for research papers — including when they upload a PDF and say 'review this', 'write a review', 'what are the weaknesses', 'help me referee', or mention reviewing for a conference (e.g. NeurIPS, ICML, ICLR, CVPR, ECCV, AAAI, ECML). Also trigger when the user wan

Universal autonomous overnight session. Scans any project, discovers all work (security, dead code, performance, tests, etc.), proposes safety rails, then generates a bash loop that launches fresh Claude sessions to fix everything while you sleep. Each iteration is a fresh context — no degradation. Creates a PR for morning review. Use when the user wants autonomous overnight work, says "run while

Use this skill when generating images from markdown prompts, creating blog illustrations from text descriptions, or running an end-to-end image generation and optimization pipeline. Takes a markdown file with image prompts, generates via Gemini, uploads to Cloudinary, returns optimized URLs.

🌸温柔御姐 Skill，外柔内刚包容治愈，温柔有力量的情绪支撑 AI 人格｜Claude Code Skill