🔍代码审查 / GitHub MCP 推荐

3D-Review-Ergebnis als PDF-Bericht erzeugen: Zusammenfassung, Tabellen, Risikoampeln. Normen: §§ 174 ff. InsO. Prüfraster: Vollständigkeit Berichtinhalte, Layout, Signaturfeld. Output: PDF-Bericht 3D-Tabellenreview. Abgrenzung: nicht Excel-Export.

Used to review whether an Apache ShardingSphere PR truly fixes the root cause, assess side effects and regression risks, and determine whether it can be safely merged. If not mergeable, produce committer-tone change request suggestions. Supports targeted comparison across GitHub-visible review rounds when prior PR comments or review threads exist. Before final output, internally self-iterate the review until no new actionable findings are discovered.

Convert markdown files to professional PDF documents with proper Chinese font support, theme system, and visual self-check. Use whenever the user asks to create PDFs, convert markdown to PDF, generate printable documents, or needs documents formatted for print or mobile reading. This skill MUST be used instead of manual pandoc/Chrome invocations — it handles CJK typography, Chrome header/footer suppression, and mandatory visual verification that manual approaches miss. **Scope — markdown → PDF only.** For Word (.docx) output use `minimax-docx`; this skill does not produce docx and the two pipelines are intentionally orthogonal.

Use when adding or running tests, checking snapshots, choosing between Bun tests and Playwright, running lint/format/typecheck, or deciding the verification scope for a change.

Use when adding or changing packages/apps, TypeScript configs, workspace dependencies, package references, exports, or monorepo project-reference relationships.

Event-triggered and schedule-triggered automations with natural-language conditions. Use when creating automations, wiring events, or understanding how triggers fire.

Review a PR against the Pascal architectural rules — package boundaries (core/viewer/editor/nodes), the registry-driven composition model (def.geometry / def.renderer / def.system), legacy-dispatch regressions, hook hygiene (useEditor/useScene/useViewer), and selector performance. Use when the user asks to review a PR, audit a branch, or check that changes respect the codebase's architecture.

Write tests, add test coverage, unit test, or add missing tests for Bitwarden iOS. Use when asked to "write tests", "add test coverage", "test this", "unit test", "add tests for", "missing tests", or when creating test files for new implementations.

Use when creating new pages, components, modules, or features in the orchestration cluster webapp at webapp/client/apps/orchestration-cluster-webapp/. Use when adding routes, data loading, forms, API integration, or UI components. Trigger whenever someone is building or modifying frontend feature code in webapp/client/, even for small changes like adding a column, filter, or panel.

Use when writing, modifying, or debugging unit tests in the orchestration cluster webapp at webapp/client/apps/orchestration-cluster-webapp/. Use when working with Vitest browser mode, MSW mocking, vitest-browser-react rendering, or any *.test.tsx file in the OC webapp's src/ directory. Trigger whenever someone needs to create, fix, or understand a frontend unit test in webapp/client/.

Use when migrating, porting, rewriting, or moving frontend code from operate/client/ or tasklist/client/ to the orchestration cluster webapp at webapp/client/apps/orchestration-cluster-webapp/. Trigger whenever someone mentions migrating a legacy page, component, or module to the new unified frontend, converting React Router to TanStack Router, replacing MobX stores with TanStack Query or URL state, rewriting styled-components as SCSS modules, or converting legacy test patterns to Vitest browser mode. Also use when someone asks how a legacy pattern maps to the new architecture, even for small questions like "how would I write this Operate component in the new app?" or "what's the equivalent of this Tasklist store in the unified frontend?"

Editable cells for `@tanstack/react-table` v9 via `@tanstack/react-form`. The table is the layout primitive; the form owns editing state. Use `createFormHook` to register reusable field components (`TextField`, `NumberField`, `SelectField`), then in each column's `cell` return `<form.AppField name={`data[${row.index}].field`}>{(field) => <field.TextField />}</form.AppField>`. Critical typing gotcha: if your row has a recursive `subRows`, use `Omit<Row, 'subRows'>` for the form row type — TanStack Form's `DeepKeys` recurses and hits TS2589. Subscribe to `form.state.values.data.length` (not the whole array) for row add/remove re-renders.

Convert a client-side `@tanstack/react-table` v9 table to server-side (manual modes). Pass server-paginated/sorted/filtered rows as `data`, set `manualPagination` / `manualSorting` / `manualFiltering` / `manualGrouping` / `manualExpanding` for whatever the server now owns, supply `rowCount` so `getPageCount()` works, and DROP the matching `rowModels` entry (no `paginatedRowModel` if the server paginates). Own the relevant state slices via external atoms (`useCreateAtom` + `options.atoms`) so a query can key on the slice and refetch automatically — OR via classic `state` + `on*Change` controlled state.

Ship-ready optimizations for `@tanstack/react-table` v9: tree-shake the bundle by registering ONLY the `features` you actually use; memoize `features`, `data`, and `columns` for stable identity; replace `(state) => state` with narrow selectors or per-slice `useSelector(table.atoms.<slice>)` subscriptions; and push state-driven re-renders down the tree with `<table.Subscribe>` / `<Subscribe>` so the expensive table body doesn't re-render every time you toggle a sort indicator. Don't over-optimize small tables — the default selector + inline rendering is fine until measured perf demands more.

Wiring reactivity for `@tanstack/react-table` v9. Covers `useTable` (and its second-argument selector), reading state via `table.state` / `table.store` / `table.atoms.<slice>`, rendering with `table.FlexRender`, opting subtrees into fine-grained reactivity with `<table.Subscribe>` and the standalone `<Subscribe>`, owning slices with external atoms via `useCreateAtom` + `options.atoms`, and packaging shared config into a reusable hook with `createTableHook` (`useAppTable`, `createAppColumnHelper`, `table.AppTable` / `table.AppHeader` / `table.AppCell` / `table.AppFooter`). Routing keywords: useTable, useSelector, useCreateAtom, atoms, react-store, table.Subscribe, FlexRender.

Wire up TanStack Devtools for TanStack Table in React. Mount `TanStackDevtools` with `tableDevtoolsPlugin()` once at the app root and call `useTanStackTableDevtools(table)` after each `useTable` so the table is registered as a devtools target. Live devtools are tree-shaken to no-ops in production unless you import from `@tanstack/react-table-devtools/production`.

Mechanical breaking-change migration from `@tanstack/react-table` v8 to v9. Every v8-shaped option, type, or method an agent will reproduce from muscle memory has a v9 equivalent enumerated below: `useReactTable` → `useTable`, root `get*RowModel` options → `rowModels` with factory + *Fns parameter, `createColumnHelper<TData>` → `createColumnHelper<typeof features, TData>`, `table.getState()` → `table.state` / `table.store.state` / `table.atoms.X.get()`, `sortingFn` → `sortFn`, `enablePinning` → split, `_`-prefixed APIs unprefixed, `ColumnSizing` split into `columnSizingFeature` + `columnResizingFeature`. For incremental migration, `useLegacyTable` from `@tanstack/react-table/legacy` accepts the v8 API on the v9 engine — deprecated, larger bundle, no `table.Subscribe`. Long-term you migrate every table off it.

End-to-end first-table journey for `@tanstack/react-table` v9. Install the React adapter, declare `features` via `tableFeatures()`, declare `rowModels` factories with their *Fns parameters (`createSortedRowModel(sortFns)` etc.), create a column helper with both `TFeatures` and `TData` generics, instantiate `useTable`, and render with `<table.FlexRender>`. New users land here, not on `useLegacyTable`.

`@tanstack/react-table` v9 does NOT include virtualization — pair with `@tanstack/react-virtual`. Standard row-virtualization pattern: get the row array from `table.getRowModel().rows`, feed `rows.length` to `useVirtualizer({ count, estimateSize, getScrollElement, ... })` in the DEEPEST possible component (a `TableBody`, NOT `App`), iterate `rowVirtualizer.getVirtualItems()` instead of `rows.map`, absolute-position each row with `transform: translateY(virtualRow.start)px`, and render `<tbody>` as a CSS grid with a fixed total height. Column virtualization uses `horizontal: true` plus padding-left/right placeholder cells. An experimental ref-mutation variant skips React reconciliation for ~10% extra perf but the standard pattern is the default.

`@tanstack/react-table` v9 is built on TanStack Store. Each state slice (sorting, pagination, rowSelection, columnFilters, …) is a separate atom. The table exposes three READ surfaces — `table.atoms.<slice>` (per-slice readonly), `table.store` (flat readonly view), `table.state` (selector output from `useTable`) — and two WRITE paths — internal `table.baseAtoms.<slice>` OR YOUR `options.atoms[slice]` if you opt to own the slice. Use `useCreateAtom` from `@tanstack/react-store` for stable identity, `useSelector` for fine-grained reads, and pass the atom in `options.atoms` so the table writes through it directly — no `on*Change` handler required.

Server-side / async data flow for `@tanstack/react-table` v9 with `@tanstack/react-query`. Canonical pattern: external pagination atom via `useCreateAtom<PaginationState>` + `options.atoms` (NOT `state + on*Change`), pagination object as part of `queryKey`, `manualPagination: true`, `placeholderData: keepPreviousData` to avoid the 0-rows flash, and `defaultData = useMemo(() => [], [])` to keep `data` reference stable between fetches. `rowCount` from the API response so `getPageCount()` works.

Use when an app developer wants to build an entity (a.k.a. an agent) for their Electric Agents app — designing a single entity type, picking a coordination pattern when needed (single-agent, manager-worker, pipeline, map-reduce, dispatcher, blackboard, reactive-observer), defining state, handler, schemas, and implementing it in one entity file. Applies to any use of `registry.define(...)` / `defineEntity(...)` in a `@electric-ax/agents-runtime` app.

Guides the design and structuring of workflow-based Claude Code skills with multi-step phases, decision trees, subagent delegation, and progressive disclosure. Use when creating skills that involve sequential pipelines, routing patterns, safety gates, task tracking, phased execution, or any multi-step workflow. Also applies when reviewing or refactoring existing workflow skills for quality.

Meta-skill that analyzes the Trail of Bits Testing Handbook (appsec.guide) and generates Claude Code skills for security testing tools and techniques. Use when creating new skills based on handbook content.

Testing practices for this monorepo — choosing between unit and end-to-end tests, where test files live, how tsconfig.test.json fits in, how to drive Playwright, to to write/adapt UI/end-to-end tests, and which test runner (vitest vs jest) to use. MUST be invoked BEFORE any code search or file reads whenever the user asks to add, write, or extend a test, add coverage, create a test file, set up testing in a new package, mentions vitest/jest, or refers to an existing `*.test.ts` file. Applies equally to unit tests and to end-to-end / UI tests — the skill routes to a Playwright-MCP-first discovery flow for the latter, which differs from normal code-search workflow.

Reviews GitHub pull requests for the Medusa repository. Checks PR template compliance, contribution guidelines, code conventions, security, performance, and bugs. Emits a structured review decision (labels + review template) for a downstream deterministic step to apply. Use when a PR is opened or updated.

Use when building, changing, or refactoring large Langfuse frontend features, virtualized lists, large tables, controller components, local feature state, Zustand stores, row selection, high-frequency UI state, or rendering-performance issues.

创建或修改 Cat Café skill / MCP tool description 的元技能（含质量标准、范本、发布）。 Use when: 写新 skill、修改现有 skill、写/改 MCP tool description、验证 skill 质量； 或者功能实现中产出了 SKILL.md / cat-cafe-skills/ 新目录 / manifest.yaml skill 条目。 Not for: 使用 skill（直接触发对应 skill）。 Output: 新/更新的 SKILL.md + manifest 条目 + symlinks。 GOTCHA: 软硬同重——skill/MCP 质量 = 代码质量；不要写模型已知的通用教程，先过价值门禁。

Internal Codex-flavored copy of the backend UI design guidance for `@open-mercato/ui`. Prefer the main `om-backend-ui-design` skill for normal backend page, CRUD, table, and form work.

Provides security review capability for TypeScript/Node.js applications, validates code against XSS, injection, CSRF, JWT/OAuth2 flaws, dependency CVEs, and secrets exposure. Use when performing security audits, before deployment, reviewing authentication/authorization implementations, or ensuring OWASP compliance for Express, NestJS, and Next.js. Triggers on "security review", "check for security issues", "TypeScript security audit".

Review requirements or plan documents using parallel persona agents that surface role-specific issues. Use when a requirements document or plan document exists and the user wants to improve it.

Mobile specialist for Flutter, React Native, and Swift native iOS development. Use for mobile app, Flutter, Dart, React Native, Swift, SwiftUI, iOS, Android, Riverpod, swift-openapi-generator, and widget work.

Generate comprehensive eval tests for any MCP server using @mcpjam/sdk. Supports Jest and Vitest with deterministic and LLM-driven test patterns.

Prevent Kubernetes hallucinations by diagnosing and fixing failure modes: insecure workload defaults, resource starvation, network exposure, privilege sprawl, fragile rollouts, and API drift. Use when generating, reviewing, refactoring, or migrating manifests, Helm charts, Kustomize overlays, cluster policies, and platform-specific Kubernetes work for EKS, GKE, AKS, OpenShift, GitOps controllers, or observability stacks.

Design cross-national survey experiments: power, equivalence, localization.

Design survey instruments: questions, scales, flow, social desirability.

Design conjoint experiments: attributes, power, AMCE/AMIE estimation.

Use when adding or changing a resource on the Huma-backed /api/v2 API (new endpoints, porting a v1 resource, editing pkg/routes/api/v2/). Covers per-operation Huma handlers, the shared envelopes, error/auth bridging, REST verb conventions, and what's automatic.

MUST USE whenever a task needs a commit or git-history investigation. Covers atomic commits, staging, commit-message style, rebase, squash, fixup/autosquash, blame, bisect, reflog, git log -S/-G, and questions like who wrote this or when was this added. Do not use for ordinary code edits unless the user asks for git work.

根据风格描述从内置设计系统库中选取最匹配的设计系统，提供 DESIGN.md 规范供后续设计工作使用。

Draft MongoDB MCP Server release notes for a new version by fetching the corresponding release from mongodb-js/mongodb-mcp-server and formatting them for the docs. Use when asked to draft or create MCP Server release notes.

Review code changes adversarially using parallel review layers (Blind Hunter, Edge Case Hunter, Acceptance Auditor) with structured triage into actionable categories. Use when the user says "run code review" or "review this code"

Analyze and create a testing strategy for native Android apps - install testing libraries, set up test infrastructure, create harnesses for unit tests, UI tests, screenshot tests, and end-to-end tests.

Inspect UI via cursor-ide-browser (Cursor Browser tab); fall back to user-playwright if internal browser MCP is unavailable. Use for @Browser, open tab, layout checks, localhost — or when user explicitly asks for Playwright.

将 MCP Server 的工具列表转换为 Skill 格式的结构化 Markdown 文档

PDF 文档解析。自动区分文字型 PDF 与扫描型 PDF，覆盖：文本/表格提取、多页全量扫描、嵌入图表 caption、单位感知数值计算。

Write step-by-step automation instructions for a workflow, tailored to your tool (Claude, n8n or Zapier). Best run right after discover-patterns or process-analyst.

FastGPT API 开发规范。重点强调使用 zod schema 定义入参和出参,在 API 文档中声明路由信息,编写对应的 OpenAPI 文档,以及在 API 路由中使用 schema.parse 进行验证。