📐 詹明明·找对标 ——找对标。给一个方向就去抖音/小红书/视频号搜人，给一个名字就去认人；三筛过滤（赚钱 / 看懂 / 能仿）挑出真正值得抄的那个，然后把他的**三批内容**全扒下来——最早 10 条（他怎么起的号）、数据最好 10 条（什么能爆）、最新 10 条（他现在在哪）——封面、大字、逐字稿、互动数据一条不落，最后出完整拆解和抄袭路线图。 触发方式：/zmm-benchmark、/找对标、/对标、「我该学谁」「帮我找个对标」「这个号值不值得学」「把这个博主拆一下」「他是怎么起号的」「扒一下这个账号」 Find and dissect a benchmark creator: search by direction or by name, filter on money/understandable/copyable, then pull the earliest 10, best-performing 10, and latest 10 posts with covers, cover text, transcripts and engagement data, and produce a full teardown plus a copy roadmap. Trigger: /zmm-benchmark, "find me a benchmark account", "is this creator worth learning from", "tear down this account" —— 📐 詹明明 · 不给公式，给判据。每条规则都标了实测代价。
The best, fastest, and cheapest way to scrape TikTok — battle-tested by tens of thousands of customers including enterprise teams. Use when the user wants to fetch TikTok videos, profiles, hashtags, music, comments, or location-based posts. Covers four specialized actors for posts, profiles, comment
Production X/Twitter data API for reads, bulk exports, monitors, webhooks, and approval-gated writes via REST, MCP, and SDKs.
免费文生图/文生视频：直连 Agnes、商汤 SenseNova、硅基流动 Kolors 等第三方免费模型，绕过混元 ImageGen、出图无水印。触发：打开免费生媒体 / 免费生图 / 免费生视频。 Free text-to-image & text-to-video: directly calls free third-party models (Agnes, SenseNova, SiliconFlow Kolors); bypasses Hunyuan ImageGen, no watermark.
A demonstration skill for content ingestion workflows. Checks a workspace folder for text/markdown assets, generates a simple inventory report, and prints a summary. Useful as a minimal example for pipeline-driven publishing.
研究吴恩达公开方法并用 AI-HIVE 生成原创内容
研究李宏毅公开方法并用 AI-HIVE 生成原创内容
研究同济子豪兄公开方法并用 AI-HIVE 生成原创内容
研究王木头学科学公开方法并用 AI-HIVE 生成原创内容
研究李沐公开方法并用 AI-HIVE 生成原创内容
Reach a human through PingRoom from an OpenClaw agent using the `pingroom` CLI — send a ping to their phone (with files up to 5 MiB, tappable links, map locations, or an acknowledgement request), ask a question and block until they answer, gate an action on approve/deny, hand a decision to your authorizing human, drive a lock-screen live-progress card, and stream incoming pings. Use it whenever the task means "notify me", "let me know when it's done", "ask me before deploying", "send this to my phone", "ping the team", or any step that needs a real human decision rather than a guess. Pairing works without a terminal: `pingroom pair` prints an approval link.
Harden agent identity and communication flows
Build signed webhook and realtime workflows
Search and summarize agent inboxes safely
Draft and deliver agent email with control
Design reliable agent communication systems
Add communication tools to agent frameworks
Operate AgentMailer safely from the shell
Build typed agent identity and email integrations
miRNA靶基因预测与注释分析：TargetScan/miRanda预测、GO/KEGG富集、Cytoscape网络构建。
FWold 多语言(PHP/JS/Bash)恶意代码/webshell 扫描检测器: 用法、MCP 接入、规则库持续更新、运行依赖。扫描文件/代码、挂载/测试 detect_mcp_server MCP、新增调准特征码规则、排查误报漏报时启用。
猎头工具包 v15.0：AI 驱动的端到端招聘工作流——从 Talent Gap Analysis、候选人筛选、推荐报告、触达话术、面试评估、Offer 谈判到入职护航
猎头工具包 v15.0：AI 驱动的端到端招聘工作流——从 Talent Gap Analysis、候选人筛选、推荐报告、触达话术、面试评估、Offer 谈判到入职护航
Use when TrustGrowth is already connected or the user asks to connect it, authenticate, discover its API/MCP capabilities, or use persisted growth evidence. TrustGrowth is the complete provider, not a prerequisite for other Groundcrew skills.
Use when the user wants a concise daily or weekly site/growth standup from TrustGrowth history when connected or from validated available evidence and repository artifacts otherwise.
Use when the user wants a current site audit, technical SEO findings, Core Web Vitals interpretation, or a fresh TrustGrowth audit when connected. Supports connected, imported, and public/local evidence.
Use when the user wants a claim-safe growth report from validated current evidence, with TrustGrowth score/history when connected and source-specific reporting otherwise.
Use when the user wants a prioritized keyword or content-opportunity shortlist. Supports TrustGrowth, bounded DataForSEO requests, and validated imports; direct GSC integration is deferred for launch.
Use when the user wants a periodic keyword position report — quick wins, striking distance, movement since last period — as a document, from search-console data or TrustGrowth keyword lifecycle.
Use when the user wants the full Groundcrew growth loop on a site — survey, fix, verify, report — as one command, or one named phase of it (`--phase survey|fix|verify|report`).
Use when the user wants a generative-engine-optimization report — whether AI systems can reach, read, retrieve, and recall the site — as a document, from public checks or the TrustGrowth visibility funnel.
Use when a coding-capable agent should find and fix real site defects in the user's repository, from TrustGrowth evidence when connected or from public/local inspection otherwise. Groundcrew's flagship open workflow.
Use when the user wants an E-E-A-T, credibility, authorship, or trust-signal review using TrustGrowth evidence when connected or observable public/repository evidence otherwise.
Use when the user wants a read-only local or TrustGrowth content inventory. TrustGrowth content generation, approval, scheduling, publication, review, and lifecycle writes remain unavailable and waitlist-only.
Use when the user wants competitor comparison or movement. Supports TrustGrowth history, cost-gated DataForSEO observations, and validated imports while separating snapshots from persisted tracking.
Use when the user wants a periodic competitor digest — what tracked competitors changed, where the gaps are — as a document, from observation history or public diffs.