Do work in an Upstash Box, a sandboxed cloud container driven through the remote Upstash MCP server (mcp.upstash.com), instead of on the local machine. Use when the user asks to run, build, test, clone, or edit something remotely, in a sandbox, in the cloud, or in a box, when the deliverable is a pull request, a public preview URL, or a screenshot of a running app, when the local machine cannot deliver (no GitHub login for gh, no way to expose a port, a dirty or slow local checkout), when several independent tasks should run in parallel on separate machines, a code factory that turns a list of tasks into a list of pull requests, or when the deliverable is a video, screen recording, demo or timelapse of a browser, web app, terminal program or agent run. Applies whenever the session has the box_* and blob_* MCP tools, even when Upstash is not named.

Drive an Upstash Box (a remote sandboxed workspace) from the terminal with the `box` CLI. Use when asked to run commands, edit files, clone repos, run builds or tests, publish a public URL, browse or screenshot a page, open a pull request or issue with a screenshot attached, schedule recurring work, run an AI agent, or do any work inside a box rather than on this machine.

Use this skill when the user asks what information a Stripe Connect connected account must provide for verification, onboarding, KYC, or account requirements; when they need to compare requirements between connected-account setups; or when they ask which verification fields, documents, or business details are required for a particular platform country, account country, business type, dashboard, service agreement, or capability.

Use this skill when the user asks what information a Stripe Connect connected account must provide for verification, onboarding, KYC, or account requirements; when they need to compare requirements between connected-account setups; or when they ask which verification fields, documents, or business details are required for a particular platform country, account country, business type, dashboard, service agreement, or capability.

Designs and reviews detections from telemetry prerequisites, correlation logic, Sigma metadata, severity, status, and benign cases. Use for detection rules, Sigma review, analytic coverage, or tuning; not generic logging advice or live deployment.

Update the guide's threat intelligence through AgentSec Triage, research new coding-agent and MCP security advisories or researcher IOC lists (malicious repositories, hashes, campaign gists), and synchronize the security feed. Use for threat database refreshes or mise a jour de la base de menaces, not repository security scans or Claude Code release tracking.

Workflow 1.5: Bridge between idea discovery and auto review. Reads EXPERIMENT_PLAN.md, implements experiment code, deploys to GPU, collects initial results. Use when user says "实现实验", "implement experiments", "bridge", "从计划到跑实验", "deploy the plan", or has an experiment plan ready to execute.

Use this skill for WeChat Mini Program AI via wx.cloud.extend.AI (小程序, wx.cloud apps). Covers generateText and streamText with callbacks (onText, onEvent, onFinish); streamText needs a data wrapper, generateText returns the raw response. Also covers multimodal image input (image_url content blocks, OCR and form extraction) — vision-capable models only. Models via wx.cloud.extend.AI.createModel with groups cloudbase (main managed, also 小程序成长计划 with `hy4-preview` / `hy3` / `deepseek-v4.1-flash`), the legacy hunyuan-v3 / hunyuan-exp, or custom-*; model id goes in the `model` field. MUST run two-step preflight before code — see body. NOT for browser/Web (use ai-model-web), Node.js backend (use ai-model-nodejs), or image generation (use ai-model-nodejs).

Use this skill for WeChat Mini Program AI via wx.cloud.extend.AI (小程序, wx.cloud apps). Covers generateText and streamText with callbacks (onText, onEvent, onFinish); streamText needs a data wrapper, generateText returns the raw response. Also covers multimodal image input (image_url content blocks, OCR and form extraction) — vision-capable models only. Models via wx.cloud.extend.AI.createModel with groups cloudbase (main managed, also 小程序成长计划 with `hy4-preview` / `hy3` / `deepseek-v4.1-flash`), the legacy hunyuan-v3 / hunyuan-exp, or custom-*; model id goes in the `model` field. MUST run two-step preflight before code — see body. NOT for browser/Web (use ai-model-web), Node.js backend (use ai-model-nodejs), or image generation (use ai-model-nodejs).

Use this skill for WeChat Mini Program AI via wx.cloud.extend.AI (小程序, wx.cloud apps). Covers generateText and streamText with callbacks (onText, onEvent, onFinish); streamText needs a data wrapper, generateText returns the raw response. Also covers multimodal image input (image_url content blocks, OCR and form extraction) — vision-capable models only. Models via wx.cloud.extend.AI.createModel with groups cloudbase (main managed, also 小程序成长计划 with `hy4-preview` / `hy3` / `deepseek-v4.1-flash`), the legacy hunyuan-v3 / hunyuan-exp, or custom-*; model id goes in the `model` field. MUST run two-step preflight before code — see body. NOT for browser/Web (use ai-model-web), Node.js backend (use ai-model-nodejs), or image generation (use ai-model-nodejs).

Read academic papers from a bare arXiv abs/pdf/html URL, a PDF, a paper title, or Zotero. Use for "read this paper", "analyze this paper", "读一下 XXX", "读一下这篇", "帮我读", "看一下这篇论文", "论文笔记", quick summaries, critiques, and Zotero collection reading. A standalone arXiv link or ordinary reading request defaults to full reading and automatic note saving in the configured Obsidian/Markdown library. Honor explicit depth, destination, and no-save requests; a citation used for another task does not trigger reading.

Searches, installs, updates, and lists skills from the Antigravity Skill Vault. Use when choosing or managing workspace or global skills through the npm CLI.

Lightweight outcome tagging for meetings — won, lost, stalled, great, or noise. Use whenever the user says "tag this meeting", "mark that as a win", "that one was a loss", "tag yesterday's call as stalled", "mark this great", "that meeting was noise", "label that meeting", or any time they describe a meeting outcome in passing. Tagging takes 5 seconds and unlocks /minutes-mirror correlation analysis — the more meetings get tagged, the smarter mirror gets at telling the user what behavior patterns lead to wins. Surface this skill any time the user mentions a meeting result, win, loss, or wasted time.

Search past meeting transcripts and voice memos for specific topics, people, decisions, or ideas. Use this whenever the user asks "what did we discuss about X", "find that meeting where we talked about Y", "what did Alex say", "did we decide on", "what was that idea about", or any question that could be answered by searching their meeting history. Also use for "do I have any notes about" or "check my meetings for".

Draft user-facing Minutes release notes for a version from the commit range, recent GitHub releases, and the repository release checks. Use when the user asks to write, generate, prepare, revise, or review release notes or a changelog for a Minutes version.

Full-lifecycle blog engine with 31 sub-skills, 12 templates, 100-point scoring, and 5 agents. Routes requests to the right sub-skill: writing, rewriting, analysis, outlines, audits, schema, charts, images, repurposing, AI citation SEO, FLOW prompts, topic clusters, and multilingual publishing. Optimized for Google rankings, E-E-A-T, and AI citations. Supports any platform. Use when the user asks to create, revise, analyze, plan, translate, localize, audit, or repurpose blog content, or invokes a /blog subcommand. Do not activate merely because the word "blog" appears in an unrelated coding, theme-debugging, database, log-analysis, or definition request.

Build a blog-level strategy for a niche or business: audience segments, 3-5 content pillars with hub-and-spoke clusters, competitive gaps, measurement plan, and a 90-day roadmap. Use when the user says "blog strategy", "content strategy", "blog positioning", "content pillars", "what should I blog about". Not for a dated publishing schedule (blog-calendar), SERP-validated cluster maps (blog-cluster), or a brief or outline for one post (blog-brief, blog-outline). Makes no promise of rankings or AI citations.

Full-site blog health assessment scanning all blog files for quality scores, orphan pages, topic cannibalization, stale content, and AI citation readiness. Runs canonical batch analysis before site-wide checks. Produces per-post scores and a prioritized action queue. Use when the user asks for a site-wide blog health check across many posts ("blog audit", "blog health", "audit all posts", "site audit of the blog"). For one post, or a quality score with fixes for a few named files, use blog-analyze instead.

Score one blog post (or a folder of posts) on a 5-category 100-point scale covering content quality, SEO, E-E-A-T, technical elements, and AI citation readiness. Runs the deterministic analyzer first, then adds judgment and prioritized Critical/High/Medium/Low fixes. Advisory style diagnostics never infer authorship or affect the score. Exports markdown, JSON or table; works with MDX, markdown, HTML or a URL. Use when the user asks to analyze, score, rate or check the quality of specific posts ("analyze blog", "blog score", "check blog quality", "rate this post"). For a site-wide health check (orphan pages, cannibalization, stale content, schema across all posts) use blog-audit instead.

Detect keyword cannibalization across blog posts by extracting primary keywords from titles and headings, clustering semantically similar targets, and flagging posts competing for the same search intent. Supports local-only mode (grep-based) and DataForSEO API mode (Page Intersection endpoint at ~$0.01/call). Outputs severity-scored report with merge or differentiate recommendations. Use when user says "cannibalization", "keyword overlap", "competing pages", "duplicate keywords", "cannibalize".