Explorar skills
69.576 skills encontradas
x-req
需求与开发准备 skill。把确认后的需求直接写成可执行 task 包,并以 Q0-Q3 risk 驱动统一开发流程。 触发场景:“帮我处理需求”、“梳理需求”、“开个 task”、“新建任务”、“这个功能怎么做”、“帮我拆一下”、`x-req`,以及用户提供需求文档路径或描述预计超过 2 小时的功能。
chatbot-review
Review and improve AI chatbot responses using conversation logs from [Botsonic](https://composio.dev/toolkits/botsonic), [Docsbot](https://composio.dev/toolkits/docsbot-ai), or [Landbot](https://composio.dev/toolkits/landbot)
design-to-code
Lanhu to Code workflow for Claude Code. Convert Lanhu design URLs into Vue/React frontend code with component reuse, design token mapping, TODO-safe business logic, and optional code review.
support-metrics
Generate a support metrics summary from [Gorgias](https://composio.dev/toolkits/gorgias) tickets and optionally push to [Google Sheets](https://composio.dev/toolkits/googlesheets)
cortivex-cross-repo
Cross-repository intelligence layer that transfers learned insights between repositories through technology fingerprinting, pattern matching, and privacy-safe sharing
Implement a story according to the requirements and tasks
Implement story.md requirements and tasks.
feature-radar-scan
Discover new feature opportunities from creative brainstorming, user feedback, ecosystem trends, and cross-project research. Writes results to .feature-radar/opportunities/. MUST use this skill when the user wants to GENERATE new ideas — not evaluate existing ones. Trigger on any request to brainstorm, explore, discover, or find new feature ideas, even casual ones like "I wonder what else we could
weekly-digest-synthesizer
Compile weekly status digests from multiple .md/.txt files. Extracts project updates, action items, and blockers into a structured report. Use when aggregating team status notes into a weekly summary. Triggers: 'compile weekly digest', 'weekly digest', 'составь дайджест', 'скомпилируй статусы'.
vitest
Vitest 테스트 환경 설정 및 사용 규칙. vitest.config.ts, vi.* API, *.test.ts 파일 구조.
eval-layer
Adiciona avaliação baseada em rubrica a um código de agente existente. Ideal para avaliar agentes, medir sua qualidade ou configurar pontuação LLM-as-a-judge, suportando comparações de agente único e múltiplos sujeitos.
ariff-claude-plugins
65 plugins that turn Claude Code into an autonomous development team. 24 agents, 34 skills, 5 hooks. Includes 12-plugin anti-hallucination suite. One-line install.
code-gen
Extract Lanhu design HTML, CSS, and assets through Lanhu MCP for Claude Code design-to-code workflows. Called by design-to-code.