⌨️Cursor 必备 MCP 组合推荐

Read-only delegate to cursor-agent (Composer) for questions, investigation, design discussion, and adversarial sanity checks. No worktree, no cherry-pick, no Lead diff review — cursor-agent is locked to ask mode and cannot write. Use when user says: ask cursor, cursor sanity check, get a second opinion, adversarial review, design discussion, investigate with cursor, cursor:ask. Do NOT load for: implementation, refactor, file edits, commit/push work, anything requiring write access (use cursor:do or breezing --cursor instead).

Configure and verify the Cursor backend for Claude Code Harness. Use when user invokes cursor:setup, wants Cursor as the local default implementation backend, or asks to check Cursor plugin/agent readiness. Distribution default remains opt-in; only local env/user settings are changed when explicitly requested.

Run a Cursor Composer review as an advisory second opinion while keeping the primary review verdict on the host brain. Use when user invokes cursor:review, asks Cursor to review, or wants composer to sanity-check a diff. Cursor never owns APPROVE/REQUEST_CHANGES.

Manage MCP (Model Context Protocol) servers for openclacky: add, list, probe, remove, reconfigure. Edits ~/.clacky/mcp.json so the user never writes JSON by hand. Trigger on: add mcp, install mcp, setup mcp, configure mcp, mcp list, mcp remove, mcp probe, mcp reconfigure.

Improve or write prose: tighten, clarify, fix grammar, and adjust tone/length while preserving the author's voice and meaning. Use when users ask to edit, proofread, rewrite, polish, shorten, or change the tone of text, or to draft something from notes. Triggers on mentions of edit, proofread, rewrite, polish, tighten, make it shorter/clearer/more formal, fix grammar, 润色, 改写, 校对, 修改, 精简, 改语气, 帮我写.

Write, run, and explain SQL queries, and do quick local analysis of CSV/SQLite data with the sqlite3 CLI. Use when users want a SQL query written or fixed, want to query a .db/.sqlite file, or want to analyze a CSV with SQL (joins, group by, window functions). Triggers on mentions of SQL, query, SELECT, JOIN, GROUP BY, sqlite, database table, 查询, 数据库, 写个 sql, 联表, 分组.

Design effective tests and decide what to test: unit vs integration, edge cases, the arrange-act-assert shape, and writing the minimal failing test for a bug. Use when users want tests for code, ask what cases to cover, want to improve coverage meaningfully, or practice TDD. Triggers on mentions of test, unit test, integration test, coverage, edge cases, TDD, assert, mock, 测试, 单元测试, 用例, 覆盖率, 边界条件.

Design the domain model for the Stitch SDK. Use when mapping MCP tools to domain classes and bindings in domain-map.json. This is Stage 2 of the generation pipeline.

Verify code changes by identifying correctness, regression, security, and performance risks from diffs or patches, then produce prioritized findings with file/line evidence and concrete fixes. Use when reviewing commits, PRs, and merged patches before/after release.

Use when controlling web pages with the OpenClaw browser tool, especially multi-step flows, login checks, tab management, or recovery from stale refs/timeouts.

Creates and rewrites Flue API reference documentation from package-visible TypeScript exports and audited source-adjacent JSDoc. Use when authoring concise reference pages under apps/docs/src/content/docs/api/ or reference pages such as reference/configuration.md. Do not use for narrative guides, tutorials, or automated maintenance checks.

Reviewer for converting unit tests to browser tests for Project Bedrock. Invoke this when the user needs to review their work while removing complex Browser dependencies from tests.

ECC 的证据优先自动化清单和重叠审计工作流。当用户想要在修复任何内容之前知道哪些作业、钩子、连接器、MCP 服务器或包装器是活动的、损坏的、冗余的或缺失时使用。

Conventions and review steps for writing analytics SQL against the warehouse. Use whenever the task involves querying tables, building a report, or aggregating metrics.

Producing small, single-concern, reviewable changes. Use when refactoring, cleaning up, restructuring, decomposing, or preparing a change for PR, especially in this multi-contributor AI-assisted codebase that is prone to sprawl diffs. Covers one-concern-per-commit, extracting helpers without widening scope, not bundling unrelated edits, decomposing oversized activities, comment hygiene, and a pre-handoff self-review checklist.

Install and start MobileVC — a Claude Code mobile workspace launcher that lets the user run Claude Code on a phone (iOS / Android) with their dev machine as the backend. Use this skill when the user asks to: - "install mobilevc" / "set up mobilevc" / "装一个 mobilevc" - "在手机上用 Claude Code" / "want to use Claude Code on my phone" - "把 Claude Code 装到手机" / "mobile claude code" - "扫码连手机" 这类与 mobilevc 关联的请求 This skill installs the published npm package `@justprove/mobilevc`, runs `mobilevc start` (which prints a LAN QR code), and points the user to https://www.mobilevc.top to install the iOS / Android client. iOS installs through the TestFlight link on the homepage; Android installs through the APK download link. Skip if the user asks to MODIFY MobileVC source code or rebuild it from scratch — this skill is a launcher, not a dev environment bootstrapper.

Turn any MCP server, OpenAPI spec, or GraphQL endpoint into a CLI. Use this skill when the user wants to interact with an MCP server, OpenAPI/REST API, or GraphQL API via command line, discover available tools/endpoints, call API operations, or generate a new skill from an API. Triggers include "mcp2cli", "call this MCP server", "use this API", "list tools from", "create a skill for this API", "graphql", or any task involving MCP tool invocation, OpenAPI endpoint calls, or GraphQL queries without writing code.

CLI output formatting standards for worktrunk. Load before editing any code that calls warning_message, hint_message, error_message, info_message, eprintln, or println, or that produces strings the user will see (CLI help, progress UI, snapshot text). Documents ANSI color nesting rules, message patterns, and output system architecture.

Use for complex Agent Zero Browser form workflows involving selects, checkboxes, radios, file uploads, contenteditable fields, multi-step validation, or visually verified submission.

Use for complex Agent Zero browser automation, including multi-tab browsing, screenshots, forms, uploads, raw pointer/keyboard actions, host-vs-container browser mode, and visual verification workflows.

Ensures UI feature work ships with desktop and mobile parity, responsive behavior, and mobile Playwright E2E coverage. Use when implementing, planning, reviewing, or testing any new feature, page, component, workflow, form, dialog, sidebar, navigation, dashboard, or visual UI change; if work touches frontend or user-facing UI, this skill must run even when user mentions only desktop or says "new feature".

Android APK pentest workflow — apktool/jadx static, Frida dynamic instrumentation, SSL pinning bypass, root detection bypass, intent fuzzing, keystore extraction.

Design, debug, and harden GitHub Actions CI/CD workflows, including reusable workflows, matrix builds, self-hosted runners, OIDC authentication, caching, environments, secrets, and release automation.

Complete guide to implementing GitOps workflows with ArgoCD and Flux for automated Kubernetes deployments.

MaaNTE Python 自定义动作（CustomAction）编写指南。覆盖 agent/custom/action/ 下的 Python 代码的架构、注册、命名、maafocus 用户消息、日志、Controller API、Pipeline 集成、错误处理等编码规范和模式参考。在编写、修改或审查 Python 自定义动作，或需要了解 agent 项目结构与 MaaFramework Python 绑定集成方式时使用。

Deploy Juicebox integrations. Trigger: "deploy juicebox", "juicebox production deploy".

Audit and fix visual hierarchy, spacing, color, and depth in web UIs. Use when the user mentions "my UI looks off", "fix the design", "Tailwind styling", "color palette", "visual hierarchy", "design system", "spacing scale", or "component styling". Also trigger when building consistent design tokens, creating dark mode themes, improving data visualization clarity, or polishing UI details before launch. Covers grayscale-first workflow, constrained design scales, shadows, and component styling. For typeface selection, see web-typography. For usability audits, see ux-heuristics.

Optimize Cursor IDE performance: reduce memory usage, speed up indexing, tune AI features, and manage extensions for large codebases. Triggers on "cursor performance", "cursor slow", "cursor optimization", "cursor memory", "speed up cursor", "cursor lag".

Debug AI suggestion quality, context issues, and code generation problems in Cursor. Triggers on "debug cursor ai", "cursor suggestions wrong", "bad cursor completion", "cursor ai debug", "cursor hallucination".

Design native iOS interfaces following Apple Human Interface Guidelines. Use when the user mentions "iPhone app", "iPad layout", "SwiftUI", "UIKit", "Dynamic Island", "safe areas", "HIG compliance", "SF Symbols", "haptic feedback", or "iOS accessibility". Also trigger when building tab bars, navigation stacks, sheets, or modals for iOS, implementing dark mode, or adapting layouts for different screen sizes. Covers navigation patterns, accessibility, SF Symbols, and platform conventions. For general UI polish, see refactoring-ui. For affordance design, see design-everyday-things.

Deploy Instantly.ai webhook receivers and API integrations to cloud platforms. Use when deploying to Vercel, Cloud Run, or Fly.io, or setting up production webhook endpoints. Trigger with phrases like "deploy instantly", "instantly cloud run", "instantly vercel", "instantly webhook deployment", "instantly production deploy".

Deploy Glean custom connectors as scheduled jobs on Cloud Run, Lambda, or Fly.io. Trigger: "deploy glean connector", "glean connector hosting", "schedule glean indexing".

Deploy Deepgram integrations to production environments. Use when deploying to cloud platforms, configuring containers, or setting up Deepgram in Docker/Kubernetes/serverless. Trigger: "deploy deepgram", "deepgram docker", "deepgram kubernetes", "deepgram production deploy", "deepgram cloud run", "deepgram lambda".

Deploy and orchestrate Vertex AI ADK agents using A2A protocol. Manages AgentCard discovery, task submission, Code Execution Sandbox, and Memory Bank. Use when asked to "deploy ADK agent" or "orchestrate agents". Trigger with phrases like 'deploy', 'infrastructure', or 'CI/CD'.

Deploy applications to Kubernetes with production-ready manifests. Supports Deployments, Services, Ingress, HPA, ConfigMaps, Secrets, StatefulSets, and NetworkPolicies. Includes health checks, resource limits, auto-scaling, and TLS termination. Use when working with creating kubernetes deployments. Trigger with 'creating', 'kubernetes', 'deployments'.

Deploy inference services on CoreWeave with Helm charts and Kustomize. Use when deploying multi-model inference, managing GPU deployments at scale, or templating CoreWeave manifests. Trigger with phrases like "deploy coreweave", "coreweave helm", "coreweave kustomize", "coreweave deployment patterns".

Deploy and manage Supabase projects in production. Covers database migrations, Edge Functions deployment, secrets management, zero-downtime rollouts, blue/green branching, rollback procedures, and post-deploy health checks. Use when deploying Supabase to production, running migrations, deploying Edge Functions, managing secrets, or implementing zero-downtime deployments. Trigger: "deploy supabase", "supabase migration push", "deploy edge function", "supabase rollback", "supabase blue green", "supabase health check".

Deploy Shopify apps to Vercel, Fly.io, Railway, and Cloud Run with proper environment configuration. Use when deploying Shopify-powered applications to production, configuring platform-specific secrets, or setting up hosting. Trigger with phrases like "deploy shopify", "shopify hosting", "shopify Vercel", "shopify production deploy", "shopify Fly.io".

Deploy Replit apps with Autoscale, Reserved VM, and Static deployment types. Use when deploying to production, configuring deployment settings, setting up custom domains, or managing deployment secrets and health checks. Trigger with phrases like "deploy replit", "replit deployment", "replit autoscale", "replit reserved VM", "replit static deploy", "replit custom domain".

Deploy MaintainX integrations to production environments. Use when deploying to cloud platforms, configuring production environments, or automating deployment pipelines for MaintainX integrations. Trigger with phrases like "deploy maintainx", "maintainx deployment", "maintainx cloud deploy", "maintainx kubernetes", "maintainx docker".

Deploy Lokalise integrations to Vercel, Netlify, and Cloud Run platforms. Use when deploying apps with Lokalise translations to production, configuring platform-specific secrets, or setting up deployment pipelines. Trigger with phrases like "deploy lokalise", "lokalise Vercel", "lokalise production deploy", "lokalise Netlify", "lokalise Cloud Run".

Deploy Linear-integrated applications and track deployments. Use when deploying to production, linking deploys to issues, or setting up deployment tracking with Vercel/Railway/Cloud Run. Trigger: "deploy linear integration", "linear deployment", "linear vercel", "track linear deployments", "linear deploy tracking".

Deploy applications that integrate with Lindy AI agents. Use when deploying webhook receivers, callback handlers, or applications connected to Lindy agents. Trigger with phrases like "deploy lindy", "lindy deployment", "lindy production deploy", "release lindy integration".

Deploy a LangChain 1.0 / LangGraph 1.0 app to Cloud Run, Vercel, or LangServe correctly — timeouts sized for chain length, cold-start mitigation, SSE anti-buffering headers, Secret Manager over `.env`. Use when prepping first prod deploy, debugging a stream that hangs behind a proxy, or diagnosing p99 latency spikes. Trigger with "langchain deploy", "langchain cloud run", "langchain vercel python", "langchain langserve", "langchain docker".

Execute mobile app testing on iOS and Android devices/simulators. Use when performing specialized testing. Trigger with phrases like "test mobile app", "run iOS tests", or "validate Android functionality".

Fill any PDF locally and place each value precisely in a visual editor. Use when the user wants to fill out a PDF form, enter data into a PDF, complete a tax/insurance/bank form, or position text on a flat/scanned PDF. Handles flat (field-less) PDFs, per-character (comb) fields, and native AcroForm fields; leaves the signature blank for the user to sign.

Per-change refactoring mechanics: count callers exactly, decide inline vs keep, collapse duplicate switch branches, route raw access through a single typed boundary, surgical one-change-per-commit, post-refactor straggler sweep for stale JSDoc and dead exports. Use when actually editing code to clean it up, when deciding whether a 1-caller helper earns its keep, or when planning a sequence of small refactor commits. For the smell catalog use code-audit; for session-long simplification use collapse-pass.

Audit SkiaSharp's native dependencies for security vulnerabilities and CVEs, including Component Governance (CG) alerts from the SkiaSharp-Native and SkiaSharp Azure DevOps pipelines. Read-only investigation that produces a status report with recommendations. Use when user asks to: - Audit security issues or CVEs - Check CVE status across dependencies - Find security-related issues and their PR coverage - Get an overview of open vulnerabilities - See what security work is pending - Check Component Governance alerts - Review CG alerts from the native build pipeline Triggers: "security audit", "audit CVEs", "CVE status", "what security issues are open", "check vulnerability status", "security overview", "what CVEs need fixing", "CG alerts", "component governance", "check container CVEs". This skill is READ-ONLY. To actually fix issues, use the `native-dependency-update` skill.