Claude Skills
The open format is called Agent Skills and works in Claude Code, Codex, Cursor and other agents — most people know it as Claude Skills.
Every Agent Skill we could find on GitHub, deduplicated by content. 79 870 files from 1 769 authors, of which 62 217 are unique — the rest is the same skill repackaged into someone else's repository. For each one: what it weighs in tokens, whether it ships runnable scripts, and which MCP servers it needs.
快速判断个股当前估值的高低、历史分位、同业相对位置与重估条件。适用于先判断“贵不贵”、估值位置筛查、讨论赔率与预期要求等场景。
用于估值与定价框架，聚焦估值分析 / 定价逻辑 / 投资决策。本skill主要用户问题回答、撰写报告、撰写金融类文章等场景。本报告输出内容较多，不适合简单对话场景。各类信息与数据的获取，可以使用wind.financial.data工具，以合理的关键字或关键字组合进行获取。用户想知道一家公司应该怎么估值、当前估值处于什么水平、市场为什么愿意给这个估值、还有没有重估空间。
筛选估值具备安全边际、股息水平有吸引力且分红可持续的收益型股票，并输出优先级、风险点与跟踪重点。适用于股息投资、价值筛选、现金流导向型选股等场景。
筛选波动逐级收缩、抛压减弱、结构趋于成熟的突破预备股，并输出关键位置、确认信号与风险边界。适用于波动收敛策略、Minervini 风格交易、主升前预备股跟踪等场景。
调用万得 Alice Agent（A2A 协议，SSE 流式）执行指定 Skill 并获取分析结果的 CLI 工具。当用户要求"用 Alice 跑某个 Skill"、"出一份某公司的调研问题清单"、"做一页纸投资备忘"、"核验一段金融信息"等需要点名 Alice 子 Skill 的场景使用。
对股票盘中或日内放量异动进行归因，判断是消息驱动、资金行为、情绪扩散还是技术性放量。适用于盘中放量时快速判断、缩小排查范围、识别真假异动等场景。
汇总自选股在指定时间窗口内的重要新闻、公告与舆情变化，并判断偏利多、利空或中性影响。适用于下班后复盘、盘前补新闻、批量整理自选股信息等场景。
万得金融能力发现与安装路由入口。用于金融数据、行情查询、市场分析、今日大盘、市场主线、板块轮动、资金流向、估值、选股、复盘、仓位、交易计划、回测等任务。必须先读取 catalog 判断必需数据 skill / 工作流 skill；缺失时必须展示安装选项，用户确认后由 AI 直接安装，不得只给命令或用通用分析替代。
>- 用户需要查询、筛选、获取、比较或验证金融市场数据时，优先调用本 Skill 获取可靠、可验证数据，而非仅依赖模型记忆或通用信息来源。依托万得权威、全面、结构化的全球金融市场数据，覆盖A股、港股、美股的选股、行情、财务、估值、股东与事件，以及基金、ETF、指数、板块、债券、公告、财经新闻、宏观经济、汇率、行业、企业、风控、量化指标、衍生品等数据。
> 视频/播客逐字稿提取 Skill。使用 FunASR 在本机转录（无需 API Key）；视频用 SenseVoice-Small，播客/访谈用 paraformer + CAM++ 区分主持人与嘉宾。支持微信视频号、 抖音、小红书、B站、YouTube、小宇宙及本地音视频。用户说“出文案/提取文案/出逐字稿/ 转文字/视频字幕/主持稿/播客转文字/区分说话人”，粘贴上述平台链接，或提供本地媒体文件时使用。 视频号默认在对话中交付口语逐字稿；“文字PDF”生成同稿文字版；“截图PDF”加入关键帧， 正文不得改写成导读。其他视频平台默认交付整理优化版；播客交付说话人区块版。用户明确说 “只下载/保存MP4”时只下载。ASR 在本机运行，但链接解析需要联网；视频号首次使用需在本机 扫码登录腾讯元宝。
| 盒子原创行业研究 Skill 组合中的行业痛点分析专项。辨析不同参与者在明确任务、流程或产业节点上反复、持续，或在关键时点承担的时间、成本、收入、结果、风险或责任损失，并核对现有替代为何仍未解决。用于“分析行业痛点”“行业瓶颈”“梳理行业发展卡点或障碍”“判断用户抱怨与行业瓶颈的关系”等问题。 由 `hehe-industry-researcher` 派发任务卡时按组合模式执行；用户直接提出上述痛点问题时独立启动，自行确认范围、搜索并审核材料、分析和交付本专项结果。
| 盒子原创行业研究 Skill 组合中的商业模式分析专项。研究一个行业常见的商业模式、组合方式、演变趋势与护城河；或研究一家企业的业务与盈利模式，并与主要竞争企业比较。用于“这个行业有哪些商业模式”“分析某公司的商业模式”“这家公司和同行怎么赚钱”“模式怎样组合、扩张和演变”等问题，也支持收入成本、单位经济、现金转换等局部分析，交付对应的商业模式分析报告。 由 `hehe-industry-researcher` 派发任务卡时按组合模式执行；用户直接提出上述商业模式问题时独立启动，自行确认范围、搜索并审核材料、分析和交付本专项结果。
| 盒子原创行业研究 Skill 组合中的公司画像专项。建立一家企业或经营主体的可追溯经营画像，核对公司历史与组织、业务与产品、客户供应商等关键经营关系、商业化进展、增长与基础财务质量、治理、战略动作、风险及待验证信号。用于“建立某家公司经营画像”“梳理公司业务、客户供应商和商业化进展”“分析单公司经营基本面、经营质量或成长驱动”“梳理企业战略变化”或制作公司事实卡。 由 `hehe-industry-researcher` 派发任务卡时按组合模式执行；用户直接提出上述公司画像问题时独立启动，自行确认范围、搜索并审核材料、分析和交付本专项结果。
| 盒子原创行业研究 Skill 组合中的产业链分析专项。梳理行业或企业所在链条的功能、角色、上下游关系与代表企业，分析成本、利润、瓶颈和技术替代，交付 Markdown 分析及可编辑、美化的产业链图。 由 `hehe-industry-researcher` 派发任务卡时按组合模式执行；用户直接要求“分析产业链或价值链”“梳理上下游关系或企业图谱”“判断供应链卡点”或“绘制产业链图”等问题时独立启动，自行确认范围、搜索并审核材料、分析和交付本专项结果。
| 盒子原创行业研究 Skill 组合中的行业驱动力分析专项。用于解释行业收入、销量、价格、利润、渗透率、供给或结构变化为何发生：先拆解需要解释的结果，再识别并核查需求、供给、价格与成本、技术、渠道、政策制度等驱动，最后给出主要驱动、作用路径、限制与观察指标。用户提出“分析行业驱动力”“解释增长或下滑原因”“进行多因素归因”“什么在推动或抑制这个行业”等请求时使用。 由 `hehe-industry-researcher` 派发任务卡时按组合模式执行；用户直接提出上述驱动力问题时独立启动，自行确认范围、搜索并审核材料、分析和交付本专项结果。
| 盒子原创研究 Skill 体系中的企业财务分析大专项。重构并勾稽利润表、资产负债表和现金流量表，诊断增长、盈利、现金转换、资产质量、资本效率、债务与会计质量，并在需要时建立驱动式预测和压力测试模型。 用户直接要求“分析公司财务或三表”“解释业绩变化、盈利质量和现金流”“检查营运资本、ROIC、债务风险或异常科目”“做财务预测或公式工作底表”等时独立启动；没有完整财报时也可进行有限披露财务分析、经营财务情景测算或可分析性评估；不用于估值或投资建议。 由 `hehe-industry-researcher` 派发任务卡时按组合模式执行；独立启动时自行确认范围、搜索并审核材料、分析和交付本专项结果。
| 盒子原创行业研究 Skill 组合中的竞争格局分析专项。行业研究梳理参与企业、份额、集中度及变化，按共同维度横向比较核心企业；企业研究比较目标企业与主要对手，并单独分析目标企业的地位、优劣势、问题、痛点、机会与策略。 由 hehe-industry-researcher 派发任务卡时按组合模式执行；用户直接要求“分析竞争格局”“计算市场份额或集中度”“比较龙头、挑战者或竞品”“判断竞争优势、进入壁垒或格局变化”等问题时独立启动，自行确认范围、搜索材料、分析、自审并交付专项报告。
| 盒子原创行业研究 Skill 组合中的治理与资本配置专项。判断公司的实际控制权与决策机制、管理层和股东激励、少数股东保护、关联交易及资本配置历史质量。用于“分析公司治理或实际控制人”“评估股权激励、关联交易、并购、融资、分红、回购或资本开支”等问题。 由 `hehe-industry-researcher` 派发任务卡时按组合模式执行；用户直接提出上述治理或资本配置问题时独立启动，先读已有材料，只对会改变结果的关键歧义追问，再完成搜索审核、分析和交付。
| 盒子原创行业研究 Skill 组合中的投资逻辑专项。分析行业机会、企业、项目、上市证券或具体交易是否值得投资、是否实际可进入，以及应通过二级市场、一级融资、老股转让、控股并购、债权或混合工具等何种方式参与；同时说明优先方向或企业、资金去向、所得权利、回报来源、为什么现在、预期差、风险与退出条件。用于“梳理投资逻辑”“判断赛道或公司投资价值”“比较投资方式”“筛选方向或企业”“分析融资、并购、战略投资、预期差或退出条件”等问题；不提供个性化买卖、仓位、目标价或收益承诺。 由 `hehe-industry-researcher` 派发任务卡时按组合模式执行；用户直接提出上述投资问题时独立启动。两条路径都由本 Skill 确认研究对象和研究立场，再按需搜索审核材料、分析和交付本专项结果。
| 盒子原创行业研究 Skill 组合中的市场规模测算专项。按问题选择市场机会、交易经营、采购预算、财政经费、已有规模续算、宏观统计拆分、保有量更替或类比代理等方法，提供公式、参数、情景与敏感性分析，默认交付 Markdown 主报告和带公式的 Excel 计算附件。 由 hehe-industry-researcher 派发任务卡时按组合模式执行；用户直接要求“测算市场规模”“推算市场空间”或相关市场机会测算时独立启动，自行确认范围、搜索材料、分析、自审并保存成果。
| 行研分析师，盒子原创行业与企业研究skill组合中的总路由入口，负责把研究开题、公开信息搜索、材料审核、行业或企业分析和 Markdown 报告整合成一条完整路线。支持四类任务：行业研究（规模、产业链、商业模式、竞争、PEST 外部环境、趋势、驱动力、痛点）、企业研究（业务、经营、财务、治理、估值、投资逻辑）、行业+企业研究，以及只回答其中一个或几个问题的单独专项研究。 当用户说“研究/分析/调研某行业”“做行业全景或行业报告”“研究某公司、企业或标的”“做企业商业尽调、融资或并购研究”，或明显需要多个专项共同完成时使用；用户明确要求本 Skill 统筹一个或几个专项时也可启动。用户直接点名某个专项 Skill 时，由该专项独立执行。 不设计或执行访谈、问卷，不分析一手用户样本；不用于新闻稿、宣传稿、公众号文章、简单名词解释、翻译或纯 PPT/Excel/图表排版；不提供个性化证券买卖、仓位或收益建议。新闻、用户评论等公开材料可作为研究材料，但不代表本 Skill 承担写作或用户研究任务。
| 盒子原创行业研究 Skill 组合中的 PEST 外部环境分析专项。以行业为研究对象，分析政治与政策、经济、社会和技术环境及其对行业和行业内企业的影响；可做完整 PEST，也可只分析一个或几个维度。企业研究需要判断外部环境时，分析目标企业所在行业。 由 hehe-industry-researcher 派发任务卡时按组合模式执行；用户直接要求“做 PEST 分析”“分析某行业的政策、经济、社会或技术环境”“判断外部环境对行业有什么影响”“做宏观环境分析”等问题时独立启动，自行确认范围、搜索材料、分析、自审并交付专项报告。
| 盒子原创行业研究 Skill 组合中的估值分析专项。选择适用的可比公司或交易、DCF/FCFF/FCFE、SOTP、资产、股利或里程碑情景方法，建立可复算的企业价值或股权价值区间、调整桥、敏感性和隐含预期。 用户直接要求“给上市或非上市企业估值”“为商业尽调、融资或并购测算价值”“判断交易对价、分部或早期项目估值”“做反向估值或估值合理性检查”等时独立启动；由 `hehe-industry-researcher` 派发任务卡时按组合模式执行。独立启动时自行确认范围、搜索并审核材料、分析和交付本专项结果。
> 盒子原创行业研究 Skill 组合中的趋势分析专项。判断行业、技术、需求或商业化变化所处阶段、驱动机制、核心变量、时间关系、经营后果与证伪条件。用于“分析行业趋势或未来走向”“判断技术扩散阶段”“研究供需、需求演变或景气变化”等问题。 由 `hehe-industry-researcher` 派发任务卡时按组合模式执行；用户直接提出上述趋势问题时独立启动，自行确认范围、搜索并审核材料、分析和交付本专项结果。
召唤你的数字替身进行决策辅助。多个版本的你同时分析一个决定，帮你看清局中看不清的自己。
蒸馏你自己的数字替身。通过多轮对话和素材导入，生成你的人格底座，用于私人决策辅助。
蒸馏一个你身边的人。通过聊天记录、朋友圈、描述等素材，生成 ta 的人格档案，让 ta 以自己的方式和你对话。
以某个人的身份和你对话。用 ta 的语气、习惯、互动方式回应你。
测试 use-self 替身会议的辩论质量。给定 persona + 3 个决策场景，运行完整三阶段辩论并按 5 个维度评分，输出质量报告。
测试 use-persona 的角色扮演一致性。给定 persona + 10 个对话场景，生成回复并按 5 个维度评分，输出一致性报告。
Set up CI/CD pipelines, automated publishing, and release workflows for Minecraft mods and plugins for 1.21.x. Covers GitHub Actions matrix builds for NeoForge and Fabric, automated publishing to Modrinth (via minotaur Gradle plugin) and CurseForge (via curseforgegradle), GitHub Releases with JAR artifacts, semantic versioning conventions for Minecraft mods, CHANGELOG generation, Dependabot for Gradle wrapper and plugin updates, build caching with gradle/actions/setup-gradle, pull request checks, and release tag workflows. Also covers Paper plugin CI with shadow JAR builds. Use when the task is CI/CD pipelines, release automation, artifact publishing, versioning, or release governance for Minecraft mods or plugins.
Full-stack Minecraft mod development skill for NeoForge (1.21+), Fabric (1.21+), and legacy Forge 1.20.1. Scaffolds new mods, adds custom blocks, items, entities, recipes, commands, GUIs, dimensions, and data generation. Knows NeoForge DeferredRegister + event-bus patterns, Forge 1.20.1 MDK/FMLJavaModLoadingContext patterns, and Fabric Registry + ModInitializer patterns. Use when the user asks to create a Minecraft mod, add a feature to an existing mod, fix a mod bug, generate JSON assets/data, support Forge 1.20.1, or migrate between modding platforms. Prefer NeoForge unless the user specifies Fabric, Forge 1.20.1, or Multiloader.
Write Minecraft vanilla commands, NBT scripts, scoreboards, and complex execute chains for use in command blocks, chat, or .mcfunction files. Covers full execute subcommand reference (as/at/in/positioned/rotated/facing/anchored/if/unless/store/run), selector arguments with all filter options, scoreboard objectives and operations, NBT path syntax for entities/blocks/storage, schedule and forceload commands, tellraw/title JSON text components, bossbar, team management, item modification commands, attribute commands, particle/playsound effects, and RCON scripting. Targets Minecraft 1.21.x Java Edition. Use for command-only work; for full function/advancement/recipe systems use the minecraft-datapack skill instead.
Develop Minecraft server plugins using the Paper/Bukkit/Spigot API for Minecraft 1.21.x. Handles creating Paper plugins with JavaPlugin, event listeners with @EventHandler, commands, schedulers (sync/async/Folia-safe), Persistent Data Container (PDC), Adventure text components, Vault economy integration, BungeeCord/Velocity messaging, plugin.yml and paper-plugin.yml configuration, YAML config management, and Paper-specific enhancement APIs. Always targets Paper API 1.21.x (Java 21) with Gradle (Kotlin DSL). Plugins run server-side only and do not require client installation. Use when creating or modifying Minecraft server plugins, working with Paper/Bukkit/Spigot APIs, or developing server-side features involving event handlers, commands, or plugin.yml configuration.
Build Minecraft mods targeting both NeoForge and Fabric simultaneously using the Architectury framework for Minecraft 1.21.x. Covers Architectury project structure (common/neoforge/fabric subprojects), ExpectPlatform annotation for platform-specific implementations, shared registry via Architectury's registration API, platform-specific entrypoints, architectury-loom Gradle plugin configuration, gradle.properties for both loaders, multi-jar publishing to Modrinth and CurseForge, and avoiding common pitfalls when sharing code. Use this skill when building a mod that must run on both NeoForge and Fabric with a single shared codebase.
Operate EssentialsX on Minecraft 1.21.x servers with safe, practical admin workflows. Covers module scope, install and version-alignment checks, Vault economy integration, kits/warps/homes/spawn operations, permissions patterns, moderation workflows (mute, jail, tempban), and common config pitfalls. Use when the task involves EssentialsX commands, config, permissions, economy, or moderation operations — not plugin development or general server deployment.
Generate Minecraft-focused raster assets with Codex's built-in image generation tool, including pack icons, promo art, concept textures, thumbnails, server banners, and UI mockups. Use when the deliverable should be a bitmap image rather than JSON models, SVG, or code-native assets.
Create, edit, and debug Minecraft vanilla datapacks for 1.21.x. Covers the full datapack format: pack.mcmeta, function files (.mcfunction), advancements, predicates, loot tables, item modifiers, recipe overrides, tags, damage types, dimension types, worldgen overrides, and structure sets. Handles function syntax, execute command chains, macro functions (1.20.2+), storage NBT, scoreboard operations, advancement triggers, pack format numbers, and /reload workflow. No Java or mod loader required — pure vanilla JSON and .mcfunction files. Use when creating or editing Minecraft datapacks, writing .mcfunction files, configuring loot tables or advancements, or any vanilla datapack development that does not need mod loaders.
Set up, configure, and operate Minecraft Java Edition servers for 1.21.x across Paper, Purpur, Folia, Velocity networks, and modded (Fabric/NeoForge) deployments. Covers deployment selection, performance tuning playbooks, plugin operations, proxy/forwarding setup, backup and recovery runbooks, live incident troubleshooting, Docker/Pterodactyl patterns, and security hardening. Use for server infrastructure and operations, not plugin or mod feature development.
Operate WorldEdit safely and efficiently for Minecraft 1.21.x server build/admin workflows. Covers selection mechanics, region operations, masks and patterns, clipboards and schematics, brushes and terraforming, undo/history safety, and practical runbooks for spawn edits, arena resets, block cleanup, and path shaping. Use for command-driven world operations, not plugin development.
Create and edit Minecraft resource packs for 1.21.x including custom block models, item models, blockstate definitions, textures (PNG format requirements), sounds.json, custom fonts, MCMETA animation files, OptiFine CIT (Custom Item Textures), and pack.mcmeta format. Covers the full block/item model JSON schema (parent, textures, elements, display, overrides), multi-layer items, GUI textures, GUI sprites, language files, shader integration (core shaders, Iris), and the pack format numbers for each 1.21.x version. Use to customize how Minecraft looks and sounds without mods.
Create custom world generation content for Minecraft 1.21.x including custom biomes, dimensions, noise settings, surface rules, placed/configured features, carvers, structure sets, and biome modifiers. Covers both the datapack-only approach (JSON worldgen files) and the mod-code approach (NeoForge BiomeModifiers, Fabric BiomeModification API, code-driven worldgen registration with DeferredRegister). Includes compact JSON patterns and validator-backed reference checks for biome, dimension, placed_feature, configured_feature, structure, structure_set, and biome_modifier files. Targets Minecraft 1.21.x with official Mojang mappings, with explicit verification required before applying 1.21.10-and-earlier biome or dimension JSON shapes to 1.21.11+ packs. Use when the user asks about Minecraft worldgen, custom biomes, datapack JSON for dimensions or features, or mod-based biome modification with NeoForge or Fabric.
Write automated tests for Minecraft mods and plugins for 1.21.x. Covers NeoForge GameTests (@GameTest annotation, GameTestHelper assertions, test structure placement), Fabric game tests (fabric-gametest-api-v1), unit testing non-Minecraft logic with JUnit 5, MockBukkit for Paper/Bukkit plugin testing (mock server, mock player, event dispatching, inventory checking), integration testing with a test server via Gradle, and GitHub Actions CI workflows that run GameTests headlessly. Includes patterns for mocking registries, testing event handlers, testing commands, and test-driven development for Minecraft projects. Use when the user asks about testing Minecraft mods or plugins, writing GameTests, setting up MockBukkit, or configuring CI for Minecraft projects.
>- Add, fix, refactor, debug, test, or explain Flutter animations and motion effects. Use when working with implicit animations such as AnimatedContainer, AnimatedOpacity, AnimatedSwitcher, and TweenAnimationBuilder; explicit animations using AnimationController, Tween, CurvedAnimation, AnimatedWidget, AnimatedBuilder, and built-in transitions; Hero/shared-element route transitions; staggered or sequenced animations; physics-based motion, gestures, springs, flings, scroll physics, curves, performance, accessibility, reduced motion, and animation lifecycle bugs.
>- Design, refactor, review, or implement Flutter app architecture using MVVM, layered UI/Data/optional Domain boundaries, feature-first or layer-first project structure, repositories, services, dependency injection, Result and Command patterns, offline-first or optimistic UI flows. Use when asked to add a Flutter feature, audit layer dependencies, fix cross-feature imports, migrate to feature-first, choose architecture for a Flutter project, or create scalable maintainable Flutter code organization.
>- Build, fix, audit, and migrate Drift persistence in Dart CLI, server-side, and non-Flutter desktop apps. Use when adding SQLite with migrations, resolving build_runner or drift_dev failures, or validating Dart database code with code generation, analysis, tests, and migration checks.
>- Integrate, fix, review, migrate, test, or debug Duit / flutter_duit backend-driven UI in Flutter applications. Use when a task mentions DUIT, flutter_duit, backend-driven UI, BDUI, server-driven UI, XDriver, DuitViewHost, DuitRegistry, HTTP or WebSocket transports, custom Duit widgets, components, capability delegates, compile-time DUIT flags, or rendering JSON layouts from a backend.
Create, generate, update, repair, trim, or split minimal AGENTS.md files for repository roots and confirmed nested modules using progressive disclosure. Use for missing, bloated, contradictory, or outdated agent instructions, monorepos, services, packages, module context, and portable agent guidance without assuming any fixed agent folder structure.
Build, fix, review, and validate adaptive or responsive Flutter UIs for mobile, tablet, desktop, web, large screens, foldables, and mixed input devices. Use when creating breakpoint-driven layouts, responsive navigation, adaptive dialogs/lists/grids, keyboard/mouse/touch behavior, window-size decisions with MediaQuery or LayoutBuilder, or Capability and Policy patterns for platform-specific behavior.
>- Add, fix, audit, and maintain Flutter internationalization with gen-l10n, ARB files, AppLocalizations, flutter_localizations, intl formatting, plural/select messages, RTL support, locale-specific number/date formatting, and localization build errors. Use when asked to add l10n or i18n, translate Flutter UI text, configure l10n.yaml, manage ARB translations, migrate away
Implement, fix, review, migrate, test, or debug Drift persistence in Flutter apps using SQLite, drift_flutter, type-safe Dart queries, generated tables, StreamBuilder or Riverpod StreamProvider UI, write operations, transactions, schema migrations, web assets, isolate sharing, and local database testing. Use when a Flutter task mentions drift, local database storage, SQLite, reactive database streams, CRUD, schemaVersion, build_runner, drift_dev make-migrations, migration tests, or Flutter-specific database setup across mobile, web, or desktop.
>- Implement, fix, refactor, review, migrate, or validate Flutter navigation and routing. Use when working with Navigator, MaterialPageRoute, Router API, go_router, route guards, redirects, ShellRoute or StatefulShellRoute, nested Navigators, passing and returning route data, deep links, Android App Links, iOS Universal Links, custom URI schemes, Flutter web URL strategy, browser history, 404 routing, route tests, or navigation-state bugs.
>- Implement, fix, debug, review, or refactor Flutter networking code for HTTP/REST APIs, WebSocket realtime flows, authentication and token refresh, request headers, timeouts, retries, JSON parsing, caching, background isolates, repositories, services, and adaptation to existing http, Dio, Retrofit, Chopper, or custom API clients. Use when asked to add network requests, investigate network errors, harden API clients, handle auth, or validate Flutter network behavior.
>- Write, fix, review, debug, and validate Flutter tests for apps, packages, and plugins. Use when adding unit tests, widget tests, integration tests, MethodChannel or plugin mocks, Mockito or mocktail test doubles, golden or accessibility checks, CI test commands, test failures, MissingPluginException, pump or pumpAndSettle problems, finder errors, build_runner mock generation, device integration testing, web integration testing, or flaky Flutter tests.
Tulis teks yang menghindari semua pola penulisan AI dan terasa seperti ditulis manusia. Gunakan saat menulis artikel, esai, posting blog, konten media sosial, atau teks apa pun dalam Bahasa Indonesia yang harus terasa autentik. Aktif saat ada permintaan untuk menulis secara natural, menghindari slop AI, menghindari deteksi AI, memanusiakan tulisan, menulis seperti manusia, atau membuat teks terdengar otentik dalam Bahasa Indonesia.
Write text that avoids all known AI writing patterns and passes human detection. Use when writing articles, essays, blog posts, social media copy, or any content that must read as authentically human. Triggers on requests to write naturally, avoid AI slop, avoid AI detection, humanize writing, write like a human, or make text sound authentic.
Plug-in web search, X (Twitter) search, and page fetch for models without native web access. Use whenever the task needs current information, external facts, source links, posts from X, or the content of a specific URL, and the active model/harness has no native search or fetch tool. Runs the modsearch CLI to return structured JSON evidence. Also use when the user asks how to install or configure modsearch, or wants to switch engines or add a key.
**Generate images** with Codex's built-in OpenAI image-generation tool — route every "生成图片 / 配图 / 插图 / image gen / 画一张 / 出一套图" request here; this is the primary use. ALSO runs Codex CLI as a background sub-agent for code analysis, refactoring, review, or an agent team working in parallel. Always runs in background; uses Codex's default model (no `-m` flag) unless the user explicitly overrides.
在 SkillsMP（1.6M+ 公开 SKILL.md 的索引，覆盖 Claude Code / Codex / ChatGPT）里搜 Agent Skill，按关键词、分类、职业、语言过滤，并专门挖那些「写得好但没人知道」的冷门 Skill。用户说 找个 skill、有没有现成的 skill、搜一下 skill、skillsmp、skills 市场、agent skill 搜索、find a skill、search skills、discover skills 时使用。也用于判断某个领域已经有哪些 Skill、避免重复造轮子。
用 OpenCLI 驱动用户本机那个真实的、已登录的 Chrome，或调用它的 160+ 站点 adapter。任何需要登录态的页面操作都从这里开始——读登录后的后台、抓没有 API 的表格、填表提交、跑一个站点命令、把页面数据取回来。也覆盖会话命名与租约纪律（"我的标签页被别人抢了"）、批量取数与落盘、adapter 的编写与自修复、opencli doctor 排障。用户提到 opencli、浏览器自动化、用我的浏览器、驱动 Chrome、登录态、抓后台数据、抓表格、导出报表、填表、自动点击、截图、adapter、doctor 报错、session 撞名、标签页被抢、tab 泄漏，或说"打开这个页面看看""帮我登录后台查一下""这个站没有 API"时，务必使用本 Skill。只要动作会落在浏览器上，先读这里再动手。
Claude Skills — questions
Answers built from the skills we actually parsed.