发现最适合你需求的 AI 技能
Enable a Gmail out-of-office auto-reply with a custom message and date range.
Set up a new Google Tasks list with initial tasks.
Create a Google Shared Drive and add members with appropriate roles.
Распознает фотографии чеков, разбивает расходы по категориям (продукты, заправка, аптека, такси, животные и др.) и сохраняет их в базу (expenses.csv). Генерирует ежемесячные отчеты по затратам. Использует быструю и дешевую модель gemini-2.5-flash для извлечения и анализа данных.
Normalize receipts, reimbursement slips, and invoices into a clean expense
Scan, categorize, and total receipts for expenses. Use when recording purchases, categorizing spending, balancing monthly totals, forecasting budgets.
Implement and debug SenseAudio realtime Agent lifecycle APIs: list agents, invoke session, query status, and leave session (`/v1/realtime/agents`, `/invoke`, `/status`, `/leave`). Use this whenever user asks to start or manage SenseAudio conversational agents, continue dialogues, or handle room/conv/token lifecycle errors.
个人阅读管理与笔记系统。用于记录读书进度、管理书摘与思考、生成阅读报告。支持渐进式阅读法:书摘记录→每周回顾→月度串联。触发词:"开始读"、"记录书摘"、"阅读进度"、"读完"、"复习"、"本月读了什么"。
Local-first reading and distillation engine for documents and pasted text. Use whenever the user wants to read, summarize, extract key points, compare documents, understand a report, pull action items, or turn long text into a clear brief. Works locally with plain text, markdown, HTML, JSON, and other text-like files. No external sync or credentials.
Interact with Readarr (ebook/audiobook manager) via its REST API. Use when searching for books, monitoring authors for new releases, checking what's missing or wanted, triggering downloads, or managing the book library. Works alongside Calibre — Readarr acquires, Calibre organises.
Reactive Resume 开源简历构建器开发指南。使用 TanStack Start (React 19 + Vite)、PostgreSQL + Drizzle ORM、ORPC (Type-safe RPC)、Better Auth。当用户需要:(1) 本地开发环境搭建,(2) 自定义模板开发,(3) 数据库迁移管理,(4) PDF 导出配置,(5) 自部署配置,(6) 功能扩展开发。
Record real OpenClaw events and reactions for the current closed-loop workflow. Use for logging duplicate-skipped and proposal-created outcomes in public.openclaw_agent_events and public.openclaw_agent_reactions, and for updating trigger fire metadata when appropriate.