Official Skills
Featured Agent Skills
Ad Creative Engine
Turn a product or offer into a structured spread of ad creative: distinct ANGLES across families (pain, desire, social proof, contrarian, mechanism, identity, urgency), each expanded into a full ad variant (primary text, headline, CTA, visual direction), checked against the target platform's char limits and ad policy, then ranked into a prioritized TEST PLAN naming what to test first and the metric that decides the winner. Use whenever the user wants ad copy, ad creative, Facebook/Meta/Instagram ads, TikTok ads, Google/search ads, ad variations, ad angles, ad concepts, a creative test plan, or says "write me an ad", "ad variants", "ads that convert". Trigger phrases: "ad creative", "ad copy", "ad variants", "ad angles", "Meta ads", "Facebook ads", "TikTok ads", "Google ads", "paid ads", "creative testing". Do NOT use to sharpen one hook line in isolation (use hook-forge), to map the whole funnel across stages (use full-funnel-architect), or to define brand voice/colors/fonts (use brand-kit).
Ats Resume Tailor
Tailor ONE resume to ONE job posting so it mirrors the posting's language, clears ATS keyword screening, and returns a match score + a named gap list. Parses the posting's must-haves, maps the user's real experience to each, rewrites bullets as quantified impact statements using the posting's exact keywords TRUTHFULLY, then scores keyword + requirement coverage and flags what isn't evidenced. Use whenever the user wants to tailor/customize/optimize a resume for a job, beat the ATS / applicant tracking system, fix keyword screening, or asks "why am I not getting interviews" with a resume + a job description. Trigger phrases: "tailor my resume", "match this job description", "ATS", "applicant tracking", "optimize my resume for this role", "score my resume against this posting", "keyword match". Do NOT use to write a resume from scratch with no target job, to write a cover letter (use cover-letter-writer), or to prep interview answers (use interview-simulator). Never fabricates roles, dates, or results.
Advisory Council
Convene a panel of distinct expert personas to debate ONE real decision and return a decision memo with an explicit recommendation, a confidence level, AND the strongest dissent attached. Personas get divergent priors, cross-examine each other's weakest assumptions, and the skill synthesizes to a single call instead of a polite round-robin. Use whenever the user brings a decision or trade-off and wants multiple expert viewpoints, a panel, a council, a board, a devil's advocate, or pros-and-cons-with-a-verdict. Trigger phrases: "advisory council", "agent council", "ask a panel", "convene experts", "what would a CFO/skeptic/customer say", "debate this decision", "second opinion", "should I X or Y", "help me decide". Do NOT use to pressure-test an already-chosen option for failure modes (use adversarial-stress-test), to write the implementation spec (use write-spec), or to draft the stakeholder announcement of a made call (use stakeholder-update).
Market Research Report
Turn a market or topic into a structured report with explicit TAM/SAM/SOM sizing (assumptions shown, sensitivity noted, top-down AND bottom-up), buyer segments, the competitive field with white space, trends and risks, and a plain-English so-what — then verify by grading every load-bearing claim's source as primary / secondary / weak and flagging shaky figures. Use whenever the user wants to size a market, write a market-research or market-sizing report, decide whether a market is worth entering, estimate TAM/SAM/SOM, map an industry's players and segments, or get a market scan with the numbers backed up. Trigger phrases: "size this market", "market research report", "TAM/SAM/SOM", "is this market worth entering", "market scan", "competitive landscape and sizing". Do NOT use for open-ended fact-finding with no market frame (use deep-research), a head-to-head on named rivals (use competitor-battlecard), or building the financial model itself (use spreadsheet-builder).
Content Atomizer
Turn ONE long-form source (podcast, webinar, YouTube video, livestream, keynote, blog post, newsletter, or raw transcript) into a complete multi-platform content package — extracting standalone "atoms," scoring them, and rewriting each natively for its destination platform. Use whenever the user wants to repurpose, atomize, or "slice up" long content, get "20 posts from one video," turn a transcript/episode into social content, or run a one-to-many content workflow. Trigger phrases: "repurpose this", "turn this podcast/webinar/video into posts", "content from my transcript", "atomize", "one piece of content into many", "what can I post from this". Do NOT use to write a single post from scratch with no source (use hook-forge / retention-script), to plan an original multi-post series from scratch, or to analyze why content performed (use performance-teardown).
Grill Me
Conducts rigorous, structured interviews to stress-test project plans and resolve architectural ambiguities before implementation begins.
Top Agent Skills
Latest Agent Skills
Build with Better Design
Builds and redesigns web or mobile interfaces using standardized design systems, UI/UX principles, and automated component kits.
Resume Tailor & Job Match
Tailors and scores resumes against job descriptions to optimize ATS keyword matching and impact without inventing false experience.
Cold Email Domain Check
Audits cold email sending domains against critical DNS, SPF, DKIM, and DMARC standards to maximize inbox placement.
Anti Ai Designer
Research-first UI and product design that never looks AI-generated. Use when designing apps, sites, screens, flows, design systems or any interface where the user wants original, category-aware, production-ready work instead of soft-blue templates, purple gradients or card grids. Triggers on design, UI, UX, app screen, landing page, mobile, web, dashboard, redesign, make it look premium, stop looking AI, award-winning design, native iOS Android, design system, design tokens. Runs research, three competing directions, build, subtraction, then measured verification with bundled scripts.
Narodowy Bank Polski (NBP)
Fetch official NBP exchange rates, gold prices, and compute compliant Polish tax conversions for foreign currency invoices.
Voice Turn Endpoint Detector
Evaluates audio frame energies and transcript fragments to heuristically detect voice turn endpoints in conversational agent pipelines.
MCP Market Hub
Manage All Your Skills in One Place
Install, update, and sync Agent Skills across Claude, Cursor, and Codex — keep every environment in sync from a single dashboard.
