Start and control Minutes Coach, the separate real-time copilot HUD, with an explicit meeting goal. Use only for explicit Coach or HUD lifecycle requests such as "start Minutes Coach", "open the Coach HUD", "pause Minutes Coach", "resume Minutes Coach", "Minutes Coach status", or "stop Minutes Coach". Do not use for requests that explicitly ask the current terminal agent to watch or strategize; those belong to minutes-live-sidekick. An ambiguous request such as "coach me live" requires one short surface clarification and must not automatically start Coach.

Manage old recordings — find large files, archive old meetings, delete processed originals. Use when the user says "clean up recordings", "how much space are meetings using", "delete old recordings", "archive meetings", "manage meeting storage", or asks about disk space from minutes.

Create irresistible offers using the Value Equation, bonus stacking, risk-reversing guarantees, and ethical scarcity. Use when the user mentions "grand slam offer", "make my offer more compelling", "what bonuses should I add", "guarantee strategy", "offer naming", or "people say its too expensive". Also trigger when packaging a product for higher perceived value, justifying premium pricing instead of discounting, designing a money-back guarantee, or structuring tiers to maximize conversions. Covers the MAGIC naming formula and starving-crowd targeting. For product positioning, see obviously-awesome. For outbound sales, see predictable-revenue.

Talk to customers without leading them using Mom Test rules: discuss their life not your idea, ask about specifics in the past, and talk less. Use when the user mentions "customer interviews", "validate my idea", "users say they want it but dont buy", "leading questions", "The Mom Test", "customer feedback bias", or "interview script". Also trigger when preparing user-research questions, interpreting ambiguous feedback, or designing customer-discovery that avoids false positives. Covers commitment and advancement, avoiding compliments, and extracting signal from noise. For product-market fit, see jobs-to-be-done. For rapid prototype testing, see design-sprint.

Choose and audit startup metrics using Croll and Yoskovitz's "Lean Analytics". Use when the user mentions "what metrics should we track", "KPIs", "north star metric", "One Metric That Matters (OMTM)", "vanity metrics", "analytics dashboard", "DAU/MAU", "churn benchmark", or "measure product-market fit". Also trigger when choosing metrics for a startup or feature, auditing a dashboard for vanity metrics, setting metric targets and baselines, or instrumenting a product by business model and stage. Covers good-vs-vanity metrics, the One Metric That Matters, metrics by business model, the five startup stages, and benchmarks. For the build-measure-learn loop, see lean-startup. For fixing activation and retention, see improve-retention.

Run the Nemotron-3.5 Super VL finance LoRA cookbook (NeMo Megatron-Bridge) end-to-end for the user, on one 8-GPU node or a Slurm cluster: data prep, checkpoint conversion, text+vision LoRA fine-tuning of the 120B-A12B model, merge to a Hugging Face checkpoint, and a vLLM try-out. Use when the user wants to run this cookbook, fine-tune Nemotron-3.5 Super VL with LoRA, or adapt it to their hardware.

Use @data-client/vue composables for data fetching, mutations, and rendering - useSuspense, useFetch, useQuery, useCache, useLive, useDLE, useSubscription, useController, useLoading, useDebounce, DataClientPlugin, Suspense, onErrorCaptured. Use when reading/rendering remote data, triggering mutations, doing optimistic updates, real-time subscriptions, or wiring Suspense and error handling in Vue 3.

Add or optimize a tensor-level kernel in veomni/ops, register its variants, integrate it with models, and add numerical and registry tests. Trigger: 'add op', 'new kernel', 'add attention variant', 'new fused op', 'add triton kernel', 'optimize operator'.

Routes serverless requests to a compute form factor and the skill that owns it. Applies first to Lambda, API Gateway, EventBridge, Step Functions and event-driven work while the compute, orchestration or integration choice is open — API backends, webhooks, jobs, cron, queue/stream consumers, durable workflows, event routing, Step Functions vs Durable Functions, and Pipes vs native event source mappings for queues and streams. Decides Event Functions on-demand vs Managed Instances, Web Functions, MicroVMs, EventBridge rules/Scheduler/Pipes/bus, HTTP vs REST API. Owns shared Lambda references (cold starts, SnapStart, concurrency, event sources, function URLs, throttling, troubleshooting). Defers to the specialized skill once a request commits to one form factor or service, or describes its defining feature, even unnamed (durable checkpointing and automatic step retries, Firecracker sandbox, managed instances, Node.js server). Not for EC2, ECS/EKS/Fargate or Amplify hosting.

Screen US stocks for Stockbee-style short-term Momentum Burst setups using 4% breakout, dollar breakout, range expansion, volume expansion, prior range contraction, close-location, failure filters, and risk-distance scoring. Use when the user asks for Stockbee, Pradeep Bonde, momentum burst, 4% breakout, range expansion, dollar breakout, short-term swing momentum candidates, or 3-5 day burst setup review.

Image SEO audit — make a site's images discoverable in Google Images and stop them from dragging down Core Web Vitals. Audits alt text quality and coverage, descriptive file names, modern formats (WebP/AVIF), dimensions and compression, responsive srcset, lazy-loading, explicit width/height (CLS), image sitemaps, and ImageObject structured data. Use this skill whenever the user asks about image SEO, alt text, Google Images ranking, image optimization for search, image file size hurting page speed, WebP/AVIF, or image structured data. Trigger on: "image SEO", "alt text", "rank in Google Images", "optimize my images for search", "image alt tags", "WebP", "image compression SEO", "image sitemap", "ImageObject schema", or any image-discoverability / image-performance question. For overall page speed use /seo-analysis; for generating new images use the brand image skills.

Inspect generated images, video and audio before delivery using deterministic metadata, contact-sheet, black-frame and silence checks plus visual and listening review. Use to accept, reject or repair Magic Hour media outputs; it does not generate media or predict audience performance.

Verify and debug native, React Native, Expo, or Flutter apps on an iOS Simulator with agent-device. Use when an agent needs to launch an app, inspect its live UI, tap, type, scroll, validate a code change, collect failure evidence, or reproduce a workflow on an iPhone or iPad Simulator.

FIN7 (Carbon Spider / Sangria Tempest) adversary-emulation playbook — revenue-targeted spearphishing with phone follow-up, EDR-evasion tradecraft, AD compromise, and big-game-hunting ransomware. Use when emulating a high-end financially-motivated crew that graduated from POS theft to ransomware. Triggers on: 'emulate FIN7', 'Carbanak', 'Carbon Spider', 'Sangria Tempest', 'big game hunting', 'EDR evasion', 'AvNeutralizer'.

Prove one task's acceptance criteria with observable end-to-end tests against the real product, record a test plan and a test-gap ledger, and capture screenshots for the pull request. Use for the QA stage, after work and before review, when FACTORY.json enables stages.qa.

Maps the `hunkdiff/extension` authoring surface for Hunk, the terminal diff viewer — hiding or reordering reviewed files, docked panes, alternate file views, commands and key bindings, dialogs, workspace writes, themes, syntax languages, VCS backends, lifecycle events. Use when writing, debugging, or installing a Hunk extension, or when a request asks Hunk itself to behave differently. Not for reviewing a diff in a live session — that is hunk-review.

Build the site-facing parts of an EmDash CMS project on Astro, including schema and seeds, content queries, Portable Text rendering, menus, taxonomies, widgets, and deployment configuration. Use for EmDash site and theme work; use creating-plugins instead when the task is plugin authoring alone.

Run, configure, and troubleshoot OpenAI Codex CLI in non-interactive headless environments. Use for Codex automation in Bash or PowerShell, native Windows or WSL2, shell scripts, CI/CD, Docker, Kubernetes, remote servers, agent harnesses, or batch jobs; for constructing `codex exec` commands; selecting sandbox and approval modes; consuming JSONL events or structured output; resuming sessions; passing prompts through stdin; and handling failures caused by unavailable interactive input such as `request_user_input`.

Use this skill when reviewing a Highcharts pull request and the `gh` CLI is available, to fetch PR details, diff, author permission, linked issues, duplicate PRs, CI checks and the author's open PR count. Don't use it when GitHub MCP or built-in GitHub tools are available; fetch the same data with those.

Use this skill when extracting or storing Biome AST/CST source text, working with any `SyntaxNode`, `SyntaxToken`, `TokenText`, quoted strings, token-relative ranges, or embedded-language attribute values, or diagnosing trivia, allocation, range, and temporary-borrow problems. Do not use for parser implementation or formatter comment placement.