import-art
将专辑封面复制到音频和内容目录的正确位置并按配置组织文件。
▸ 推荐理由省去手动复制到多处路径的繁琐，适合批量或频繁管理封面文件的创作者。
▸ 风险提示会访问并修改本地文件路径，需注意路径与权限。
全部收录
相关性达标的全量 Skills（含未精选长尾）· 共 5090 个
help
展示插件可用技能、常见工作流与快速参考帮助信息。
▸ 推荐理由对新手友好，快速引导常用流程并列出可调用技能与命令。
health-check
运行插件健康检查，验证环境、依赖与技能注册状况。
▸ 推荐理由便捷定位插件配置与依赖问题，适合排查安装与运行异常。
genre-creator
为 bitwize-music 生成新的流派（genre）README 文件并规范化 slug。
▸ 推荐理由自动化新增流派文档并处理命名与路径，适合维护库和扩展分类。
explicit-checker
扫描歌词并核验显性内容标注是否准确。
▸ 推荐理由发布前检查脏话与标注不一致，降低被下架风险
document-hunter
用浏览器自动化检索并下载公共档案与原始文献。
▸ 推荐理由快速抓取政府/法院等公开原始文献，节省人工查找
▸ 风险提示会以自动化方式访问并下载外部网站内容，可能触及反爬或下载限制
configure
交互式设置或编辑插件配置文件并校验配置有效性。
▸ 推荐理由首次安装或调整参数时的必备管理工具
cloud-uploader
将宣传视频与专辑内容上传到 S3 或 Cloudflare R2 存储。
▸ 推荐理由便于发布与分发，直接生成可分享链接
▸ 风险提示需要提供云存储密钥并会访问外部存储域名
clipboard
将曲目内容复制到系统剪贴板以便粘贴到外部工具。
▸ 推荐理由方便将歌词或风格提示快速导出到生成工具
▸ 风险提示会执行本地剪贴板命令，需授予主机访问权限
album-ideas
管理专辑创意库：增删查改并跟踪状态。
▸ 推荐理由把零散灵感结构化为可执行的创作任务
album-dashboard
生成专辑制作进度面板，显示阶段完成率与阻塞项。
▸ 推荐理由一目了然掌握制作进度与卡点
album-conceptualizer
按七阶段流程设计专辑概念、叙事弧与曲目构架。
▸ 推荐理由快速为专辑搭建概念与叙事蓝图
album-art-director
为专辑设计视觉概念并生成针对不同平台的 AI 艺术提示。
▸ 推荐理由把音乐概念转为可直接使用的图像生成提示，节省设计时间。
▸ 风险提示生成的视觉提示可能涉及版权或敏感素材，使用时注意合规。
figma-code-sync
Keep an existing Figma design system aligned with the codebase as the code changes over time — propagating token, spacing, icon, colour or breakpoint changes into Figma and proving nothing broke. Use when code and Figma have drifted, when a design-system change in code must reach Figma (or the reverse), when auditing a Figma file against the code that ships, when deciding between variable modes and component variants, when a per-instance fix in Figma "succeeds" but nothing changes on canvas, when checking whether a Figma component's variants and props still match the code component, when checking contrast on token pairs inside the Figma file, when a Figma page or template is built from raw frames instead of component instances, when deciding which tier a component belongs to or in what order to build a component library, when a template's grid, margins or spacing need to come from tokens, or when building a DTCG token pipeline so the two sides cannot diverge. Complements figma-generate
design-system-code
Build and enforce a design system inside a codebase — token tiers that nothing can skip, and guards that stop the code from drifting off them. Works for both new and existing projects, which need different guards. Use when starting a project and setting up tokens and enforcement from the first commit, when installing design-system enforcement on an existing repo, when tokens exist but nobody uses them, when hardcoded colours/spacing/typography keep reappearing, when deciding between a ratchet and a hard rule, when a design-system guard needs writing, extending or debugging, when migrating a large codebase onto tokens without a red build from day one, or when a "we have tokens" system is not actually being consumed. The code-side counterpart of figma-generate-library and figma-code-sync.
i-have-adhd
为有 ADHD 的读者调整回复格式，突出下一步行动并保持会话持续模式。
▸ 推荐理由适合需要结构化、可执行指令的用户，提升注意力与执行力。
i-have-adhd
使代理输出适配 ADHD 读者：强调下一步、列步骤、限制跑题与估时。
▸ 推荐理由显著提升面向 ADHD 用户的可操作性与持续会话一致性。
obsidian-second-brain
让 LLM 以自我改写方式管理并维护你的 Obsidian 知识库，实现长期自治的“第二大脑”。
▸ 推荐理由实现来源重写与自动调和冲突，适合需要长期沉淀与自动维护笔记的用户。
▸ 风险提示需授予对本地 Obsidian vault 的读写权限，存在隐私与数据被改写风险。
whisper-identity
Give this agent a real, routable IPv6 (/128) identity on the Whisper network, with safe egress and externally-verifiable identity (DNSSEC + RDAP). Keyless verification needs no account.
us-english-spelling-sweep
Sweep a repository to US English, sized by blast radius — every finding reported before anything changes, and identifier renames and string edits confirmed one at a time. Use when the user asks to check spelling or grammar across a repo, wants a consistent pass after spotting a single UK spelling, or is settling a project's language before a release or handoff.
mentor
Mentor — coach while the user writes every line themselves, so they learn by doing instead of receiving a finished diff. Use when someone asks for "mentor mode" or "learning mode", tells you not to do the work for them ("teach me, don't do it for me", "let me drive", "guide me, don't take over", "don't write it, help me write it"), or is learning a language or framework and wants to type it themselves. Not for someone who wants the code explained while you still write it.
tdd
Test-driven development gated on user approval of the failing test. Use when implementing new behavior, fixing a bug, or when the user asks for TDD by any name. Not for docs, config, or refactors already covered by tests.
dep-updates-combiner
Combines multiple open dependency-update PRs (Dependabot, Renovate, or any dep-update bot) into one branch and one PR, so CI runs once instead of once per bump. Fires on an explicit ask to combine them, and equally on the bare complaint that dependency PRs have piled up or that CI reruns on every bump. Reserved for collapsing several bot PRs into one: a single PR on its own (fixing a failing bump, closing it, merging it) and configuring the bot are other work.
changeset
Changesets — write the changeset file directly, since the interactive CLI asks questions an agent cannot answer. Use when a change to released code is ready to commit and has no changeset yet, when a check fails with "packages have been changed but no changesets were found", when only docs or tooling changed and the release needs an explicit "nothing to release", or when the user asks for a changeset, a release note, or a version bump. Not for running a release or editing versions and changelogs by hand — automation owns those.
agent-config-audit
Audit a repository's agent configuration for rot — pointers to things that moved, commands that no longer run, facts that stopped being true, wiring nothing ever registered. Use when the user asks to audit the agent config, CLAUDE.md or AGENTS.md; after restructuring the config layer; or when an agent has just followed an instruction that turned out to be wrong. Reads the instruction layer alone — the files an agent consults before it acts. Reports its findings and waits for approval before changing anything.
storage-analyzer
为 macOS/Windows 做只读磁盘占用扫描并生成交互式清理报告。
▸ 推荐理由能自动定位大文件并分级给出可执行清理方案，适合磁盘空间不足时使用。
▸ 风险提示可能在本地启动服务并展示可执行删除命令，执行需用户确认。
neat-freak
审计并收尾项目文档、代码与运行态，确保现役答案一致可交接。
▸ 推荐理由减少交接成本，确保文档与实际行为一致，便于长期维护。
leader
把一句想法拆成可直接交给 agent 执行的目标任务书和验收标准。
▸ 推荐理由能产出可直接运行的 /goal 级任务书并支持分工与断点续跑，适合多 agent 项目。
khazix-writer
以卡兹克（Khazix）风格生成公众号长文、扩写与续写稿件。
▸ 推荐理由针对公众号长文有明确风格与流程，便于内容运营产出一致性稿件。
hv-analysis
用横纵双轴法对产品/公司/技术做系统性深度研究并生成 PDF 报告。
▸ 推荐理由方法论清晰，适合做竞品与历时对比并产出可阅读报告。
▸ 风险提示需安装本地 Python/WeasyPrint 等依赖以生成 PDF。
aihot
通过 AI HOT 的公开 API 提供中文 AI 资讯、热榜与日报摘要。
▸ 推荐理由实时抓取公开数据源，适合追踪 AI 圈动态与产品/论文发布。
▸ 风险提示会向外部 API (aihot.virxact.com) 发起匿名只读请求。
valuing-information
Prices research, tests, surveys, dashboards and experiments before running them, by computing how much the result would raise expected utility - which is zero whenever no outcome would change the decision. Use when someone proposes an A/B test, user survey, market study, competitor analysis, analytics build, consultant, pilot, or "let's gather more data", and when deciding how large a study needs to be.
tracking-beliefs
Separates signal from noise in metrics you cannot observe directly, updating a belief from evidence instead of reacting to the latest reading - Bayesian updates over competing explanations, and a filter that says whether this week's move is real. Use when a metric moves and someone wants to act, when diagnosing why traffic or revenue changed, when a dashboard number contradicts intuition, or when deciding whether a trend is real yet.
stress-testing-plans
Validates a plan before committing to it - checks whether the ranking survives the assumptions, finds the most likely way it fails, and makes the trade-offs explicit via a Pareto frontier instead of an invented single score. Use before a launch, price change, migration, infrastructure change or major commitment, when running a premortem, when two objectives conflict, or when a decision depends on an assumption nobody has tested.
reading-rivals
Models what a competitor will do next and what happens if you respond - best response, equilibrium, and the difference between a rival who optimises perfectly and one who does not. Use before a price change or price war, when a competitor launches something, when deciding whether to match a rival's move, when analysing a market with few players, or when someone asks "what will they do if we do this".
planning-horizons
Decides how far ahead to plan, what to discount future payoffs by, and which backlog items can be dropped without analysis - using receding-horizon planning, an explicit discount factor, and branch-and-bound pruning against the incumbent. Use when building a roadmap, prioritising a backlog, arguing about short-term versus long-term, setting quarterly or annual plans, or when planning has become an end in itself.
learning-from-outcomes
Turns past decisions into calibrated judgment - scoring old predictions against what happened, assigning credit for delayed results, and separating a bad decision from a good decision that lost. Use during a retrospective, postmortem, quarterly or monthly review, when reviewing decisions taken 30+ days ago, when attributing a result to a cause, or when someone claims to have called something.
framing-decisions
Turns a vague "what should we do about X" into a scored decision - explicit actions, an explicit unknown, an explicit prior, and one utility scale - then picks by maximum expected utility and checks the result for framing effects. Use when facing a choice between options under uncertainty, when a decision has stalled in circular debate, when someone asks "should we do A or B", or before writing any decision record, strategy memo, or roadmap commitment.
allocating-effort
Splits scarce time, budget or traffic across competing products, channels, campaigns or variants by treating them as a multi-armed bandit - Thompson sampling over beta posteriors, with optimism for anything not yet tried enough to judge. Use when deciding what to work on next, how to divide a marketing budget, which product to prioritise, whether to kill something that is underperforming, or when picking a winner among test variants.
find-complementary-founders
在隐私保护下发布临时化名资料并匹配互补的联合创始人或项目合伙人。
▸ 推荐理由适合早期创始人寻找合伙人，强调可验证的 owner-to-owner 隐私机制。
▸ 风险提示会发布可被检索的公开或半公开资料，存在身份披露或隐私泄露风险。