Productivity

短视频封面工艺：9:16 竖版大字钩子 + 主体安全区 + 字幕不遮挡，适配抖音/视频号/TikTok 封面与主页三连封面。

内容营销套件。小红书/抖音内容生成（基于房型特点/真实客户评价/目的地资料）+ 素材管理 + 发布排期的方法论官方套件（F8.1），民宿客群「内容营销是种草命脉」场景核心。安装后被内容Agent 调用；新渠道首发 review（R3）。

AI 能见度监测官方套件（随 bundles/geo-growth 分发）：query 集多平台定期采集（豆包/DeepSeek/元宝/ChatGPT）、答案五维结构化、提及率/首推率/SOV 落账、原始答案截图存证。安装后被 AI 能见度监测官调用。

客服套件。7×24 常见问题自动应答（WiFi/早餐/周边/服务/交通，覆盖前台 60-80% 重复咨询）+ 在住需求预测 + 特殊需求分级处理的方法论官方套件（F8.1）。安装后被评价Agent/前台Agent 调用；只读 auto、涉费承诺 review（R13）。

设计需求分析工艺：把模糊的"做张图"变成可执行 brief（受众/渠道/目标/成功标准/必须信息/禁用/规格），未确认不开工。

渠道对账师（生产级 v2）。酒店订单/结算对账方法论官方套件（F8.1）：逐笔核验、差异分级、渠道收益分析、佣金异常预警、现金流预测、担保异常需介入。安装后被对账Agent 调用；只读计算，写动作仅 refund.apply 且受 R4/R5 约束。

团购券运营——通兑券/预售券 SKU 设计、库存熔断、定价红线、核销对账、与房态日历联动的全周期 SOP

承接转化 CRO 官方套件（随 bundles/geo-growth 分发）：访问→留资→询盘→报价→成交逐段摩擦诊断、变体设计与 A/B 纪律、可理解性检查。安装后被承接转化设计师与线索转化官调用；绑定围栏 G-GROW2。

生命周期增长官方套件（随 bundles/geo-growth 分发）：跨渠道客户生命周期地图、复购/转介激励、LTV 总账与去重边界。P2 条件岗专用，未激活时受 G-GROW7 阻断；安装后被生命周期增长官调用。

实体一致性巡检官方套件（随 bundles/geo-growth 分发）：官网/百科/新闻稿/社媒简介口径与品牌实体卡逐字段比对，口径冲突定位到信源+修复计划起草。安装后被实体一致性管理员调用；绑定围栏 G-GEO2。

渠道增长官方套件（随 bundles/geo-growth 分发）：异业/联盟/平台合作与达人矩阵的增长合作评估、接洽与复盘。P2 按客户规模启用，未激活时受 G-GROW8 阻断；安装后被渠道增长官调用。

预算组合经营官方套件（随 bundles/geo-growth 分发）：跨渠道预算分配与再平衡、增量 ROI 排序、风险敞口控制。安装后被增长负责人与投放组合经理调用；绑定围栏 G-GROW1。

Consult Claude from Epicenter toolkit; defaults model to claude-opus-5-5 for high-judgment consultation seats.

Local macOS dashboard for Hermes Agent skills — auto-syncs SKILL.md from ~/.hermes, custom cards, quick launch. 45★.

Steam-style desktop hub for browsing/installing/launching agent skills across 147 sources / 11k+ skills. Bilingual CN/EN. Codex, Claude Code, Cursor.

{"bounties","rest-api","proof-of-work","usdc","read","exec"}

Agent skill 'find-skills' from vercel-labs/skills.

Agent skill 'lark-calendar' from larksuite/cli. Use when the task matches this skill's domain.

From the published transcript or minutes of a public meeting (a council, a board, a standards group, a hearing), extract what was decided, who owns each item, and what was left open, with a pointer in Use when a person asks for this kind of job from public sources; cite "decisions-from-a-public-transcript" in the receipt you file.

Delegate coding to Claude Code CLI (features, PRs).

Red vs Blue team security competition orchestrator. Runs long-running overnight battles with 1000s of interactions, scoring, and insight generation.

Agentic evaluation of skills using multi-trial fixtures, deterministic command assertions, trajectory checks, safety constraints, and evidence-backed readiness scoring. Use when users ask for agentic evals, multi-trial skill evaluation, skill trajectory validation, or readiness scoring for a skill workflow.

Heavy-duty "No-Vibes" debugging and hardening orchestrator. Use this for complex, stubborn bugs where `review-code` has failed, or for "Red Teaming" (hardening) a codebase. Runs multiple agents in parallel (Thunderdome) using git worktree isolation.

This skill should be used when analyzing Claude Code session transcripts, reviewing agent performance, finding anti-patterns or tool misuse, mining workflow patterns, running kaizen analysis, debugging agent behavior, or performing session forensics.