Build voice AI agents with LiveKit. Use when developing realtime voice applications, voice agent pipelines (STT-LLM-TTS), WebRTC communication, or deploying conversational AI. Covers LiveKit Agents SDK, provider selection (Deepgram, OpenAI, ElevenLabs, Car
5,032 Steam results · Checked
Stop drifting back to chatbot behavior. Patterns for maintaining autonomous agent effectiveness across sessions. 不要再回到聊天机器人的行为了。跨会话维持自主代理有效性的模式。
Generate and send a 5-day Tübingen weather graphic (PNG) from open-meteo.com. Use when Master wants a nicer visual forecast plus summary text for the next few days. 从 open-meteo.com 生成并发送 5 天蒂宾根天气图(PNG)。当主人想要更美观的可视化预测以及接下来的几天摘要文字时使用。
Send daily 08:00 weather reports for Tübingen using open-meteo.com. Use when Master wants an automated summary (current conditions + today’s high/low + rain chance) stored locally and forwarded via Telegram. 每天08:00发送T宾根的天气报告,使用open-meteo.com。当Master需要自动摘要
Use when creating new skills, editing existing skills, or verifying skills work before deployment 在创建新技能、编辑现有技能或在部署之前验证技能是否有效时使用
Use when you have a spec or requirements for a multi-step task, before touching code 当您在接触代码之前对多步骤任务有规范或要求时使用
Use when about to claim work is complete, fixed, or passing, before committing or creating PRs - requires running verification commands and confirming output before making any success claims; evidence before assertions always 在提交或创建 PR 之前,即将声明工作已完成、修复或通过时使
Weekly reset: commitments, stalled work, next-week plan. 周度复盘与计划:汇总承诺、停滞事项与下周安排,作为 cron 定时任务运行,非通用方法论。
54 real design systems (Stripe, Linear, Vercel) as HTML/CSS. 54 套真实设计系统(Stripe、Linear、Vercel 等)的 HTML/CSS 模板,完整复刻配色、字体、组件与间距体系。
Forces the laziest solution that actually works, simplest, shortest, most minimal. Channels a senior dev who has seen everything: question whether the task needs to exist at all (YAGNI), reach for the standard library before custom code, native platform fe
Turn meeting notes into cited decisions, owners, tickets. 将会议记录/转写稿转为可追责的决策、责任人与工单,支持与 Teams 会议管线及项目看板联动。
Ultra-compressed communication mode. Cuts output tokens 65% (measured) by speaking like caveman while keeping full technical accuracy. Supports intensity levels: lite, full (default), ultra, wenyan-lite, wenyan-full, wenyan-ultra. Use when user says “cavem