发现 Skills

Auto-activate for litestar_granian imports, GranianPlugin, granian CLI commands, ASGI server config in a Litestar app. The litestar-granian plugin is a zero-configuration replacement for litestar run that uses Granian (a Rust-based ASGI server) instead of uvicorn. It produces GranianPlugin configurations, GranianConfig tuning (workers, threads, HTTP, SSL, backpressure), and Litestar app lifespan integration. Use when serving a Litestar app in development or production, replacing uvicorn, configuring workers/threads, enabling HTTP/2 or SSL, or tuning backpressure. Not for FastAPI, Django, or non-Litestar apps — use the plain granian CLI for those. Not for uvicorn — Granian is preferred for all Litestar deployments.

Auto-activate for litestar.contrib.htmx imports, HTMXRequest, HTMXResponse, hx-* attributes in templates served by Litestar. Litestar's first-party HTMX integration: HTMXRequest helpers, HTMXResponse, TriggerEvent, Refresh, PushUrl, HXLocation, partial-HTML responses, OOB swaps, and pairing with litestar-vite for HTMX mode. Produces Litestar Controllers/handlers that return partial HTML, HTMX response helpers (TriggerEvent / Reswap / Retarget), and template patterns for hx-* driven UIs. Use when: building HTMX-powered Litestar apps, returning partial HTML, triggering client events from server, integrating htmx with litestar-vite. Not for full SPAs (React/Vue/Svelte — see litestar-vite SPA mode), not for non-Litestar HTMX apps.

Auto-activate for litestar_saq imports, SAQPlugin, SAQConfig, QueueConfig, TaskQueues. The first-party Litestar plugin around SAQ (Simple Async Queue): background tasks, cron jobs, queue web UI, `litestar workers run` CLI, DI of `TaskQueues`. Produces SAQPlugin configs, QueueConfig definitions, task functions, CronJobs, and DI-injected enqueue patterns. Use when: adding background jobs to a Litestar app, scheduling cron work, exposing the SAQ web UI, or running workers via the Litestar CLI. Not for Celery, RQ, or Dramatiq — Litestar's first-party choice is SAQ. For raw SAQ patterns outside Litestar, see standalone SAQ docs. Or document the custom-PG-native alternative pattern when SAQ is explicitly rejected.

Capture and document a new idea in this repo. Use when the user says things like "help me brainstorm", "I have an idea", or "let's capture this for the future" and wants it recorded in `design/ideas/` with a summary, supporting research (repo context + web if useful), and a sketch.

Web 自动化测试与浏览器控制

name: omk description: An intelligent agent for the oh-my-knowledge evaluation tool for knowledge artifacts. It evaluates the quality of skills (system prompts), compares different versions, and automatically iterates to improve them. Use when: the user mentions "evaluation", "assessment", "eval", "benchmark", "compare skill", "improve skill", "evolve", "generate test cases", "gen-samples", or "omk". user-...

Fonte canônica atual: `skills/lk-leads-esfriando/SKILL.md`.

REST API design, implementation, and best practices. Use for API endpoints, HTTP methods, status codes, authentication, and API documentation.

Runs and interprets the AI DevSecOps Policy Enforcement Agent for CI/CD and GitOps. Use when reviewing pipelines (GitLab CI, GitHub Actions), Argo CD manifests, Kubernetes configs for security/compliance, generating policy findings, remediation suggestions, or auto-fix patches. Trigger on DevSecOps, policy enforcement, pipeline security, supply chain, GitOps, Argo CD, SBOM, or compliance-aware review.

Configure npm, yarn, and pnpm to use China-based registry mirrors (npmmirror/Taobao, Tencent Cloud, TUNA, USTC) for faster package installation. Use this skill whenever npm install, yarn install, or pnpm install is slow or timing out in China, when ETIMEDOUT or ECONNREFUSED errors appear from registry.npmjs.org, or when setting up a Node.js development environment in China. If the user says "npm 太慢", "yarn 超时", or similar, always use this skill.

A structured, repeatable method for turning goals into verifiable, auditable actions, defining required inputs, preconditions, and a stepwise execution procedure.

Fetch and summarize Feishu group chat history. Use when the user asks to read, review, or summarize messages from a Feishu group chat. Triggers: "看群聊记录", "群里聊了啥", "帮我看看这个群", "群消息历史", "chat history", "what did the group discuss". NOT for: sending messages (use message tool), reading documents (use feishu-doc skill), or wiki operations (use feishu-wiki skill).

Feishu cloud storage file management. Activate when the user mentions cloud space, folders, or the drive.

Modernize legacy applications and codebases. Use for COBOL conversion, framework upgrades, and technical debt reduction.

Controls ROS/ROS2 robots via rosbridge WebSocket CLI. Use when the user asks about ROS topics, services, nodes, parameters, actions, robot movement, sensor data, or any ROS/ROS2 robot interaction.

Proactive project memory management via MemOS MCP. USE MCP TOOLS AUTOMATICALLY when: (1) Starting work - memos_search for context, (2) Completing tasks - memos_save as MILESTONE, (3) Fixing bugs - memos_save as ERROR_PATTERN, (4) Making decisions - memos_save as DECISION, (5) Encountering errors - memos_search for solutions, (6) User mentions '之前/上次/previously' - memos_search history, (7) Need to understand dependencies/causality - memos_get_graph or memos_trace_path for relationships, (8) Cube not found - memos_list_cubes to discover available cubes, (9) Need full memory details - memos_get with memory_id. Available MCP tools: memos_search, memos_search_context, memos_save, memos_list_v2, memos_get, memos_suggest, memos_list_cubes, memos_get_graph, memos_trace_path, memos_export_schema.

Generate daily and weekly codebase activity summaries from Azure DevOps (PRs, work items), local git commits, and Confluence. Use when asked about 'what happened', 'codebase activity', 'team summary', 'pr summary', or to understand what's going on in a large codebase.

Builds AI agents on Cloudflare using the Agents SDK with state management, real-time WebSockets, scheduled tasks, tool integration, and chat capabilities. Generates production-ready agent code deployed to Workers. Use when: user wants to "build an agent", "AI agent", "chat agent", "stateful agent", mentions "Agents SDK", needs "real-time AI", "WebSocket AI", or asks about agent "state management", "scheduled tasks", or "tool calling". Biases towards retrieval from Cloudflare docs over pre-trained knowledge.

Description: Track and analyze portfolio company performance against plan. Ingests monthly/quarterly financial packages (Excel, PDF), extracts KPIs, flags variances to budget, and produces summary ...

Create exercise directory structures with sections, problems, solutions, and explainers that pass linting. Use when user wants to scaffold exercises, create exercise stubs, or set up a new course section.

Scaffold and architect custom Frappe apps including app structure, hooks, background jobs, service layers, and production hardening. Use when creating new apps, setting up app architecture, or implementing cross-cutting patterns like caching, logging, and error handling.

Effect Atom v4 patterns for React state management. Use when working with effect/unstable/reactivity, atoms, AtomRegistry, useAtomValue, useAtom, Atom.family, runtime.atom, derived atoms, persistent storage. Triggers on Atom.make, Atom.writable, Atom.kvs, useAtomSuspense, AsyncResult, BrowserKeyValueStore, effect/unstable/reactivity.

Use when reviewing code changes before merge. Multi-axis review with severity labeling.

当需要为前端组件生成测试用例时使用。前端测试自动生成与执行，为UI组件和页面自动生成组件测试、视觉回归测试、E2E测试和无障碍测试，确保前端代码质量可自动化验证。关键词：前端测试、组件测试、视觉回归、E2E测试、无障碍测试、Storybook。