AI & ML

goof00 Bundle

公司法律法规及司法解释汇编（2026年版）· 离线权威条文库。收录截至2026年8月31日现行有效的公司、证券、破产领域法律、行政法规、司法解释及中国证监会现行规章共37件全文，每件附效力状态、文号、施行日期与官方来源链接。 触发条件（用户提到以下任一内容时激活本技能）： - 公司法律问题：公司法、公司章程、股东权利、股东会/董事会/监事会、股权转让、增资减资、公司决议、董监高义务、法人人格否认、股东代表诉讼、公司解散清算、企业改制 - 登记与资本：公司登记、注册资本、认缴出资期限、企业名称、财务会计报告 - 证券领域：IPO、证券发行与承销、上市公司再融资、信息披露、上市公司治理、重大资产重组、上市公司收购、要约收购、投资者关系、国有股权 - 破产领域：企业破产法、破产申请与受理、管理人、破产财产、债权申报、债权人会议、重整、和解、破产清算 - 用户要求"查公司法第X条""查破产法规定""查上市公司收购规则"等精确条文检索 本技能将： 1. 先读总索引定位主题，再按路由表**只读取相关分册**（禁止一次性读入全部） 2. 从本地汇编中精确摘录条文原文，附效力状态、文号、施行日期与官方来源链接 3. 汇编未收录的，明确说明并建议用华宇元典/北大法宝 MCP 补充检索，绝不凭记忆默写条文 4. 涉及2026年8月31日之后的问题，主动提示时效核验 5. 可输出条文检索结果、法律分析意见、合规审查清单三类成果 前置说明：本技能为**离线知识库**，无需任何外部连接即可使用，断网、无 MCP 环境下同样可用。

antonbronnfjell Bundle

Comprehensive repository naming convention for Git repos — personal, freelance/agency, and tiered-enterprise contexts — including dedicated patterns for AI agent skills, MCP servers, hooks, commands, and other agent-tooling repos. Use this skill whenever the user is naming, renaming, or scaffolding a new Git repository; asking what to call a repo; choosing a purpose/platform/repo-type/language prefix; or asking to validate an existing repo name against a convention. Also trigger when the user is creating a skill, MCP server, hook, agent definition, or other AI-agent-tooling repo and needs the right platform/repo-type keywords (e.g. claude vs agent vs mcp, skill vs mcpserver vs hook).

kaka100671 Bundle

通过影刀 Studio 内嵌的 studio-mcp（本地 HTTP MCP 端点）让外部 agent 完整操控影刀设计器——搭建/编辑可视化流程（VisualFlow）、编写代码流（CodeFlow）、运行验证、查指令文档。当用户提到影刀、ShadowBot、studio-mcp、搭流程、flow.edit_blocks、VisualFlow、CodeFlow、操作影刀设计器时使用。端口号随影刀会话变化，用前必须先运行 scripts/sync-studio-mcp.ps1 探测并写入 opencode 配置。

alephaitech Bundle

拆解参考图并把视觉手法迁移到用户产品；或从产品图/基础资料出发，检索可核验信息，策划详情页并输出逐屏生产级豆包生图执行稿。详情页只有产品图/零散身份规格且无完整合格卖点集时，先交付调研卖点大纲并经用户确认。适用于参考反推、主图套系、整套详情页和稀疏资料补全。不用于纯文案、普通修图或翻译。

tohman233 Bundle

Use ChatGPT in the Codex side browser/tab as a review agent for local artifacts. Default is packet review. MCP connector review needs ChatGPT Developer mode. Covers code, papers, drafts, Pro and High/extra-high reviewers.

samin12 Bundle

Break a bloated, monolithic agent system prompt into a lean stable core plus extracted Ratel-style skills. Use when a prompt is too long or does too many things, to extract or factor procedures into skills, decompose or split a prompt, cut per-turn token cost, or `/ratel-decompose-prompt`. Writes a markdown decomposition plan (lean core, skill inventory, per-skill SKILL definitions, registration wiring) to repo/.ratel/ without rewriting the prompt or code. Reached from /ratel-assessment's Prompt Decomposition dimension; pairs with /ratel-tune-definitions.

samin12 Bundle

Audit a codebase's tool and skill definitions against a failure-mode rubric and write a markdown tuning plan with per-definition before/after rewrites that make them retrievable by Ratel's hybrid lexical + semantic index and usable by the model. Use when the model picks the wrong tool, tools never get picked, descriptions are too long/vague/anemic, to fix near-duplicates, tighten schemas/enums/param names, clean up the tool catalog, or `/ratel-tune-definitions`. Fetches live Ratel docs; writes to repo/.ratel/ without editing tool or skill code. The fix /ratel-assessment's Definition Quality dimension routes to; runs after /ratel-decompose-prompt.

samin12 Bundle

SAM: zero-shot image segmentation via points, boxes, masks.

samin12 Bundle

Author in-repo SKILL.md files: frontmatter and structure.

anton-abyzov Bundle

Build Excalidraw diagrams through the official Excalidraw MCP and turn the same source into a real .excalidraw file on disk or an Obsidian .excalidraw.md drawing. Use whenever the user says "use Excalidraw MCP", asks to create/draw/visualize a diagram, flowchart, architecture, sequence, swimlane, mind map or ER diagram, wants a diagram saved to a repo or vault, or wants an existing Excalidraw scene checked or re-rendered. Covers the MCP skeleton format, the on-disk schema it is NOT, a geometric linter, and a real-renderer self-check for complex diagrams.

esemianczuk Bundle

Local cached POI, water, fuel, cafe, restroom, parking/start-anchor, and stop-insertion workflows for Route Generator MCP. Use when users ask to find, rank, add, compare, visualize, or plan a POI or parking start; this is the primary skill for a single non-private parking/start-anchor request.

esemianczuk Bundle

Diagnose explicit Route Generator MCP failures, missing state, contradictory tool results, duplicate artifacts, or provider/transport errors. Use only after a call failed or observed state conflicts with a prior confirmed result; do not trigger for ordinary route analysis.

esemianczuk Bundle

Core Route Generator MCP operating procedure for generating, summarizing, and following up on stored routes. Use for any route generation, best-in-area, loop, point-to-point, multi-route comparison, route-session continuity, or "tell me about this route" request.

palmier-io Skill

Generate ultra-realistic, UGC/influencer-style photos inside Palmier Pro — candid selfies, flash photos, lifestyle shots, and product-in-hand ad creative that read as genuine phone photos instead of AI-generated images. Drives Palmier Pro's get_timeline/get_media/generate_image tools directly (correct aspect ratio, model choice, reference-image handling). Use whenever the user wants UGC-style imagery, influencer selfies, "make it look like a real iPhone photo," authentic/non-AI-looking ad creative, or product photos held by a model — even if they don't say "UGC" explicitly.

tashfeenahmed Bundle

Get and manage a free public domain (yourname.makes.fyi or yourname.agentdomains.co) for an AI agent or app using the AgentDomains CLI. Use this whenever an agent builds a website or API and needs somewhere to put it, or needs a public hostname to expose a server, create a webhook URL, or get a stable address. Covers signup, claiming a name, pointing it at an IP or CNAME, getting HTTPS, forwarding (HTTP redirect) to an existing site, reverse-proxying the name to a backend over HTTPS, delegating to your own nameservers, removing a single DNS record, and closing the account again.

chukong-creator Bundle

Operate Agent OS v0.6 when delivery needs cross-Agent handoff, project learning, isolation, recovery, independent acceptance, or external-release evidence. Keep local reversible work on the normal single-Agent loop.

chukong-creator Bundle

Initialize, operate, inspect, and evolve governed Codex-Claude Code multi-agent collaboration across software projects. Use when Codex should act as product or technical director, Claude Code should implement and rework, either system should delegate to subagents or background agents, work must hand off around Codex limits, or a project needs Git baselines, isolated Agent worktrees, merge gates, observable state, activity logs, and collaboration health checks.

aldo10012 Bundle

Run any question, idea, or decision through a council of 5 AI advisors who independently analyze it, peer-review each other anonymously, and synthesize a final verdict. Based on Karpathy's LLM Council methodology. MANDATORY TRIGGERS: 'council this', 'run the council', 'war room this', 'pressure-test this', 'stress-test this', 'debate this'. STRONG TRIGGERS (use when combined with a real decision or tradeoff): 'should I X or Y', 'which option', 'what would you do', 'is this the right move', 'validate this', 'get multiple perspectives', 'I can't decide', 'I'm torn between'. Do NOT trigger on simple yes/no questions, factual lookups, or casual 'should I' without a meaningful tradeoff (e.g. 'should I use markdown' is not a council question). DO trigger when the user presents a genuine decision with stakes, multiple options, and context that suggests they want it pressure-tested from multiple angles.

templeking Bundle

通过本地 Node Client 调用财富管理远程 MCP Gateway，查询基金、股票行情、市场、黄金、财经资讯和金融知识。用户询问财富数据、基金或股票分析、市场行情、黄金报价、财经新闻或金融概念时使用；只依据 MCP 返回数据作答，不执行交易，不承诺收益。

templeking Skill

黄金产品对比 Skill。基于财富 MCP Gateway 的黄金价格工具，对不同银行积存金报价、伦敦金参考价格和汇率换算价格进行客观比较。适用于“哪家银行积存金价格更低”“今天各银行积存金报价有什么差异”等问题。第一版不负责黄金概念解释、个性化推荐、资产配置或交易决策。

templeking Skill

多基金对比分析 Skill。基于财富 MCP Gateway，对 2～5 只具体基金进行基础画像、收益表现、风险回撤、持仓风格和基金经理等维度的横向比较。适用于“这几只基金有什么区别”“对比收益、回撤和持仓”“哪只历史表现更稳”等问题。不负责基金筛选、个性化推荐、买卖决策或交易执行。

templeking Skill

盘中/盘后行情解读 Skill。基于财富 MCP Gateway 的市场行情、市场研判、RAG 资讯、优选板块、板块研判等 Tools，对市场或板块的涨跌现象进行轻量解释。适用于“今天市场为什么跌”“今天 A 股怎么样”“最近哪些板块比较强”“白酒板块最近怎么看”“美股最近怎么看”等问题。不负责买卖推荐、交易执行、持仓陪伴、KYC、资产配置或个性化投资建议。

templeking Skill

多股票对比分析 Skill。基于财富 MCP Gateway，对 2～5 只股票的主营业务、所属行业、生成式行业与竞争分析、同日行情和近期事件进行轻量横向比较。适用于“工行和建行有什么区别”“宁德时代和比亚迪业务上有什么差异”“今天哪只股票涨得更多”等问题。第一版不支持结构化财务、估值、技术面、资金流向或目标价对比。

ohnishant Bundle

Demonstrates a registered skill. Use to understand how to invoke skills via this MCP server. Trigger with 'run example skill' or 'show example'.

chrisduvillard Bundle

Use when the user wants an agent to explore an unfamiliar repository, synthesize candidate work, ask structured direction questions, generate a bounded Claude Code /goal or equivalent implementation goal, or safely prepare a doctrine-gated autonomous request.

binchitects Bundle

Use when working with Penpot designs through the Penpot MCP server - creating or editing components, design tokens, variants, boards, text, or layouts, and when inspecting or auditing a Penpot file. Provides corrected API signatures (several bundled MCP docs are wrong), documented gotchas that fail silently, and compact query patterns that avoid large result dumps.

portero-aylagas Bundle

Use this skill when the user wants to safely review, audit, refactor, or improve an application that contains AI behavior such as LLM calls, RAG, tools, agents, workflows, or model APIs, with characterization, local verification, deterministic AI integration tests, optional model evaluations, focused patches, and explicit approval gates for push, hooks, CI, or full automation.

dimon-ton Bundle

Use this skill whenever the user wants to interact with Google's Gemini CLI tool — including running AI queries, asking Gemini to analyze files or code, managing Gemini extensions, configuring MCP servers, managing skills, resuming sessions, or piping content to Gemini. Trigger on phrases like "ask gemini", "run gemini", "use gemini to", "gemini query", "query with gemini", "send this to gemini", "pipe to gemini", "gemini extensions", "gemini mcp", "gemini skills", "install gemini extension", or any time the user wants to use Gemini as an AI assistant from the terminal. Also trigger when user says things like "can gemini do X" or "have gemini look at this".

dimon-ton Skill

Locate, list, diagnose, and manage MCP servers in Claude Code. Use this skill whenever the user asks about MCP servers — even if they phrase it casually. Trigger on: "where is my mcp", "list mcp servers", "check mcp", "find mcp server", "mcp not showing", "mcp not connected", "add mcp server", "mcp server location", "why is mcp not working", "how do I add an mcp", "what mcps do I have", "show my mcp tools", or any variation of these. Also trigger when the user says an MCP server is missing, disabled, or not appearing in /mcp — they almost certainly need this skill.

templeking Skill

财经资讯摘要 Skill。基于财富 MCP Gateway 的资讯检索、市场概览、行情和黄金价格等工具，聚合市场、基金、黄金、股票相关热点信息，输出结构化新闻摘要、可能影响和后续关注点。适用于“今天市场有什么重要新闻”“总结今天财经热点”“今天黄金有什么大事”等问题。不负责深度投研、个性化推荐、持仓分析或交易决策。

templeking Skill

基金综合评测 Skill。基于财富 MCP Gateway 的 queryKypInfo、queryFundValuation、fundAnalysis 等 Tools，对单只基金进行产品画像、风险收益、回撤波动、持仓风格、行业暴露、基金经理基础信息和风险提示分析。适用于“这只基金怎么样”“这只基金风险大吗”“这只基金最近为什么跌”“张坤的这只基金还能拿吗”等单基金分析问题。不负责基金推荐、组合配置、交易执行、持仓陪伴或个性化投资建议。

templeking Skill

金融指标查询 Skill。基于财富 MCP Gateway 的公开查数类 Tools，支持基金、股票/指数、黄金、市场/板块等对象的单项或少量指标查询。适用于“005827 现在估值多少”“贵州茅台现在多少钱”“上证指数当前点位是多少”“现在金价多少”“这只基金风险等级是什么”等查数型问题。不负责行情诊断、资讯解读、投资研究、买卖推荐、交易执行、持仓陪伴、资产配置或个性化投资建议。

templeking Skill

个股综合分析 Skill。基于财富 MCP Gateway 的股票基础画像、行业分析、证券行情和资讯检索能力，对具体股票进行轻量综合研究。适用于“帮我分析宁德时代”“这家公司主要做什么”“最近为什么波动”等问题。第一版不支持结构化财务指标、估值水平、技术指标、资金流向、机构评级或目标价分析。

templeking Skill

跨资产关联导航 Skill。基于财富 MCP Gateway 的板块、基金持仓和股票画像数据，支持板块到相关基金、基金到重仓股票及行业风格、股票到行业概念等跨资产关系查询。适用于“半导体有哪些相关基金”“这只基金主要持有哪些股票”“这只股票属于什么行业”等问题。不负责全市场选股、基金推荐、指数成分查询、Beta 计算或交易决策。

templeking Skill

金融概念解释 Skill。基于财富 MCP Gateway 的 FAQ 知识库、通用金融问答和资讯检索能力，解释基金、股票、市场、黄金、资产配置等业务域中的金融概念、指标和产品规则。适用于“最大回撤是什么”“夏普率怎么看”“积存金是什么”等问题。不负责具体资产分析、实时行情查询、产品推荐或交易决策。

shuotao Bundle

驅動一個多模型分工的自動迴圈：由 Fable 擔任 orchestrator 規劃並審定升級方案，Sonnet 擔任 implementer 實際執行修改，另一支 Sonnet 擔任唯讀 observer 記錄每個動作供事後查核，唯讀 inspector-ops 在每個完成階段查核驗收；有錯就從最前端的 prompt（先 brief、再 plan）回頭修正再重跑，同一階段最多三次失敗就升級給 Fable 擔任 advisor 提出修正建議，全程 human out of the loop、所有選項採建議值，直到通過或觸發安全護欄才停下來問人。適用於有明確可自動驗收條件的多階段工作（重構、遷移、規格校對、registry 一致性），不適用於探索性或需要人類美感判斷的任務。觸發條件：loop-up、/loop-up、多模型分工、自動迴圈修正、multi-agent loop、orchestrator advisor loop、observer inspector、prompt-first correction、迴圈驗收、自動重試升級、fable orchestrator sonnet implementer。