学术
学术写作、研究与引用
research-add-items
Weizhena
向现有调研outline补充items（调研对象）。
compilar-tese
anavvanzin
Compile thesis chapters to DOCX or PDF via Pandoc. Use when the user says "compilar", "make tese", "gerar PDF", "gerar DOCX", or wants to build the thesis.
Article Writer
midhunxavier
Intelligent research paper writing assistant that supports multiple venues (IEEE, ACM, NeurIPS, etc.) and paper types. Includes capabilities for researching, writing, reviewing, coding analysis, and citation management.
maotou
zym27187
猫头的群聊人格与表达方式。基于当前项目中的 2 类本地语料 （7148 条 uin=954036195 原始消息 + 40 份“猫头”问答/梗图索引）的蒸馏， 提炼 5 个核心心智模型、7 条决策启发式和完整的表达 DNA。 用途：作为群聊锐评搭子，用猫头的视角吐槽游戏、动画、AI 工具、联机选择和群内乐子。 当用户提到「用猫头的视角」「猫头会怎么说」「猫头模式」「切到猫头」「让猫头锐评」时使用。 即使用户只是说「推荐点厕纸」「这玩意好不好玩」「这角色像不像猫头」「让浦西猫头说两句」「猫头团集合」「牛牛雷达响了」「贷款割包皮」也应触发。
research-suite
xiaguxunlian-dev
AI-Accelerated Scientific Research Assistant — 从文献检索到 Meta 分析，知识图谱构建， 全流程自动化。覆盖多源联邦检索（PubMed/arXiv/Semantic Scholar/OpenAlex）、 质量评估（RoB2/ROBINS-I/GRADE/JBI）、证据合成（PICO/证据表格/PRISMA）、 综述写作（IMRAD）以及 Phase 3 的 Meta 分析（效应量提取/森林图/异质性统计）和 知识图谱（实体关系抽取/Neo4j 导出/趋势分析）。
Convert Document
midhunxavier
Convert between LaTeX and DOCX formats. Preserves structure, citations, math equations, figures, and tables during conversion.
pds
jsstechio
Pencil Design System generator. Invoke with /pds followed by a business domain (e.g., "/pds coffee shop that sells coffee online"). Creates a complete design system with tokens, foundations, components, patterns, and optional screens in a .pen file.
submission-disclosures
pedrohcgs
Generate the submission-time disclosure block for a manuscript — the AI-use disclosure statement matched to the target journal's policy, CRediT author-contribution roles, conflict-of-interest statement, and data-availability statement. Use when the user says "AI disclosure", "disclosure statement", "do I need to disclose Claude", "CRediT roles", "conflict of interest statement", "data availability statement", or is preparing a submission package. NOT statistical-disclosure screening of restricted-data outputs — that is /disclosure-check.
dissertation_gec
Derrie-Vincent-Comp-Ling
Error-Type-Aware, CEFR-Adaptive Feedback for L2 Writing — MSc Computational Linguistics dissertation project.
article-rewriter
zuoa
深度改写中文文章的技能。用户要求“重写”“改写”“润色重构”“扩写”“翻译并改写”“改成公众号/知乎风格”“把这篇文章写得更适合发布”“优化这篇文章的表达或结构”时必须触发。支持 inline text、文件路径、URL 三种输入；支持 quick、standard、publish 三种模式；可按需要补充最新资料并生成 5 个标题候选。即使用户只说“把这篇改一下”“翻成中文后重写”“帮我整理成一篇可发布文章”，也应使用此技能。
dupe-research
dupe-com
Start AI-powered product research tasks on dupe.com. Kick off deep product research with archetype-based personas, track progress, and get comprehensive reports with product recommendations, pricing, and editorial articles. Use when the user wants to research, compare, or find the best products in a category.
bs-scope
MaxGiu67
"MVP Scoping: MoSCoW, anti-scope, milestone. Usa questa skill quando l'utente vuole definire lo scope, prioritizzare feature, o dice bs scope, scope mvp, MoSCoW, anti-scope, cosa fare prima, priorità ."
feedback-design
bfmcneill
Use when building loading states, progress indicators, success confirmations, or error messages. Covers feedback loops and patterns that feel satisfying.
ai-paper-writing
ParkJumyung
Best practices for writing AI research papers. Use when the project involves writing a research paper in AI field.
learn
alessandrobologna
Research a user-specified topic with current authoritative sources and package the findings as a reusable topic-kb skill (for example, kinesis-kb). Use when a user asks to learn about a topic, build a knowledge-base skill, or convert web research into reusable guidance.
last30days
FluxA-Agent-Payment
"Research any topic from the last 30 days across Reddit, X, and web. Uses paid APIs (~$0.03) for real engagement metrics. Use when user asks to research trends, find recommendations, get prompting techniques, or understand what people are discussing. Triggers on: /last30days [topic], research [topic], what's trending, best [X] recommendations."
tldr MCP Server — Skill Context
jbrogan
Person email and lastName are optional. Only firstName is required.
opencode-crane
AugustChaoTW
CRANE — Autonomous research assistant MCP server with 124 tools covering the full academic research lifecycle. Key trigger phrases: "paper trace" / "do paper trace" → trace_paper; "evaluate paper" / "score paper" → evaluate_paper_v2; "review paper" / "detect defects" → crane_review_full; "diagnose paper" / "diagnose section" → crane_diagnose; "submission check" / "投稿前檢查" → run_submission_check; "simulate submission" / "acceptance probability" → simulate_submission_outcome; "review pipeline" / "full review" → run_review_pipeline; "build index" / "scan paper" → build_paper_index; "semantic search" / "find similar" → semantic_search; "ask library" / "rag" → ask_library; "match journal" → match_journal_v2; "citation graph" → visualize_citations; "karpathy review" / "code review" → karpathy_review; "literature review" / "search papers" → run_pipeline(pipeline="literature-review"); "init project" / "setup" → run_pipeline(pipeline="full-setup"); "what can crane do" / "help" → crane_help(topic=...) or list_workflows().
tech-scout
ljchg12-hue
Expert technology scouting including emerging tech discovery, evaluation, and adoption recommendations
collect_interests
HowardKao-1130
Load recorded user interests from the personal library.
campaign-brief
tonyflo79
Synthesize all foundation work (Skills 01-08) into a single actionable campaign brief with coherence auditing and creative direction. Use after all foundation skills are complete and you need the master document that unlocks long-form writing. The Campaign Brief is the most complex foundation skill — it loads all 8 upstream packages, audits coherence across dimensions, detects weaknesses, recommends lead type, story type, and emotional arc, and produces threading guides for downstream skills. Produces the campaign brief with coherence scores, creative direction, and execution guidance. Trigger when users mention campaign brief, final brief, writing brief, or are ready to start writing copy. Requires all Skills 01-08 outputs.
solve-architecture-autonomy
IgorGanapolsky
Course-correct from over-planning to execution — when the agent has produced a plan, research, or options instead of doing work the user already approved, drop the analysis and execute the smallest end-to-end action now (branch, run, deploy, verify). Use when the user says "stop planning, just do it", "why are you researching, execute", "just ship it", "enough analysis", or pushes back that you keep proposing instead of acting on an already-decided task. Do NOT use when the task is genuinely undecided and needs a plan first, when the user explicitly asked for research/options/a design, or for routine work where no over-planning has occurred — acting on an unscoped or destructive task is its own failure.
cover-letter
adrasyn
Use when the user wants to write, review, or generate a cover letter for a job application. Triggers on requests to apply for jobs, draft cover letters, review existing cover letters, or customize application materials.
ashin-for-you
KuKuXia
陈信宏（五月天阿信）的思维框架与表达方式。基于6个维度、50+个一手与二手来源的深度调研， 提炼7个核心心智模型、8条决策启发式和完整的表达DNA。 用途：作为思维顾问，用阿信的视角分析问题、审视决策、提供反馈。 当用户提到「用阿信的视角」「阿信会怎么看」「阿信模式」「Ashin perspective」时使用。 即使用户只是说「帮我用阿信的角度想想」「如果阿信会怎么做」「切换到阿信」也应触发。