技能列表

🦞 AI小龙虾社区 Skill —— 为 AI小龙虾用户提供智能社交分身能力。安装后可通过自然语言命令操作社区（发帖、查看、回复），同时启动心跳机制（定时主动发帖、内容推送、互动提醒），以用户设定的「人设」在社区中保持存在感。适用场景：当用户想加入龙虾圈、配置AI分身、发布社区动态、查看社区内容、设置心跳频率、邀请好友时，应加载此 Skill。

>

Deep load testing workflow—goals and SLOs, workload modeling, scenario design, environment fidelity, execution, metrics interpretation, and bottlenecks to fixes. Use when validating capacity, before launches, or reproducing latency under stress.

Locate and utilize AI-friendly documentation index files (llms.txt, llms-full.txt) or sitemap.xml. Use when encountering documentation URLs (containing /docs/, /api-reference/, /guides/, or platform subdomains) to instantly map sites and reduce token burn.

Use this skill when the user wants to monitor LLM behavior over time and get alerted when outputs change unexpectedly. Triggers on requests like "set up LLM regression monitoring", "alert me when my prompts start behaving differently", "watch my LLM for regressions", "run behavioral tests on my AI outputs on a schedule", or "detect when my model starts drifting". Handles first-time setup, baseline capture, scheduled monitoring, and alert configuration.

Deep LLM evaluation workflow—quality dimensions, golden sets, human vs automatic metrics, regression suites, offline/online signals, and safe rollout gates for model or prompt changes. Use when shipping prompt updates, swapping models, or building eval harnesses for agents and RAG.

LlamaIndex RAG 框架助手，精通文档索引、检索增强生成、向量存储、查询引擎

直播带货脚本生成，话术设计、节奏把控、互动环节、促单技巧。适合直播主播、电商运营、MCN 机构。

英格兰传奇足球俱乐部

|

收集用户身体指标、生活习惯、既往病史和目标，生成可执行的长寿行动计划与分阶段复盘机制，并提供风险评分、自动周报/月报、保健品安全检查和每日饮食营养均衡分析（含热量缺口）。Use when users ask for longevity plans, healthy routine optimization, behavior-change schedules, supplement safety review, daily meal/nutrition analysis, calorie deficit tracking, or recurring reminders for hydration, standing breaks, sleep, and exercise.

Use when redesigning or polishing an accessible live website directly in the browser, especially when the goal is to produce high-quality before/after screenshots plus a reference patch script for a downstream code agent. If this skill is selected, do not also invoke any other design-related or browser-use-related skills for the same task.