Ready-made skills & agents for your AI tool. Pick, install, deploy - and suddenly your AI can do more.
Our skills & agents run directly in your tools
An SEO content brief that ranks on Google AND gets cited by AI — SEO + GEO in one.
SEO + GEO Briefing is an AI skill that turns a keyword into an SEO content brief built to rank on Google AND get cited in AI answers (ChatGPT, Perplexity, AI Overviews) — SEO + GEO/AEO in one: search intent, outline, answer-first block, meta tags, content gap and schema markup.
Turn rough notes into clean meeting minutes — with clear action items: who does what, by when.
Meeting Minutes is an AI skill that turns rough meeting notes or a raw transcript into clean meeting minutes: separated decisions and action items (who, what, by when), plus a ready-to-send follow-up email, Slack note or task export.
A/B test significance check — before you ship, I'll tell you the truth.
A/B Test Auditor & Planner is an AI skill for experimentation — an A/B test significance calculator and audit that checks statistical significance, sample size and sample ratio mismatch (frequentist or Bayesian) and gives a clear ship / hold / iterate verdict.
From target audience to buyer personas, messaging and ad targeting — all in one skill.
Audience Architect is an audience-research AI skill that builds buyer personas and an ideal customer profile (ICP) from your product: sharp customer segments and persona cards (jobs-to-be-done, buyer journey, negative list, confidence) plus voice-of-customer analysis, ad targeting and per-segment positioning. B2B/B2C preset included.
Honest, actionable feedback for (almost) anything — code review, design review, decisions and more, each with its proven expert method.
Review Architect is an AI review skill for honest code review, design review and constructive feedback on almost anything — one shared review engine, plus a focused lens per subject built on the proven method: Nielsen heuristics + WCAG (design), Google code-review standard + OWASP (code), pre-mortem (decisions), blameless postmortem + 5 Whys (processes), SBI (feedback). Findings ranked by severity with concrete fixes, no sugar-coating.
Stop your AI from agreeing with everything — a system prompt that kills sycophancy and turns the yes-man into an honest advisor with a backbone.
Tired of ChatGPT agreeing with everything? This system prompt stops the sycophancy and turns the people-pleasing assistant into an honest advisor: hard truth first, assumptions challenged, no empty flattery, every recommendation names its risk. Calibrated both ways — it agrees when you're right, owns its mistakes, and leaves the call to you.
Reach Inbox Zero: sorts your emails, pulls out the tasks and drafts each reply for you.
Inbox Pro is an AI skill for Inbox Zero and GTD email management: it sorts your inbox into clear categories, decides one action per email (do <2 min, delegate, reply, schedule, archive), extracts tasks, dates and waiting-for, and drafts each reply for you. Spots phishing and only drafts — nothing is sent or deleted without you.
Senior code review line by line: bugs, security, missing tests — with concrete fixes and a clear verdict.
Code Reviewer Pro is an AI code-review skill for pull requests, diffs and files — reviewing line by line like a senior reviewer. Catches bugs, security flaws (OWASP/STRIDE) and missing tests with concrete fixes, paste-ready Conventional Comments and a clear verdict: LGTM / Request Changes.
Open, fix and clean any messy CSV — broken, old or multi-file — then compare, reconcile and chart it.
CSV Whisperer is an AI data skill for tabular data that opens and fixes any broken or messy CSV: it auto-detects delimiter, encoding and column types, repairs garbled characters, profiles the data, compares and merges files, spots anomalies, and builds the right chart. Two-stage: quick triage first, then a clean read.
Plan your week realistically with time blocking — turn your to-do list into a weekly schedule with a buffer for the unexpected.
Weekly Planner is an AI planning skill that turns a messy to-do list into a realistic, time-blocked weekly schedule across one or more weeks: priorities (Eisenhower/MITs), deep work in your peak hours, breaks and ~20% buffer. Factors in your recurring workload, calibrates through reviews, and honestly says what won't fit instead of overloading you.
Write a newsletter from one topic: subject lines, preview text, structure and call-to-action — benefit-driven, not ad fluff.
Newsletter Forge is an AI newsletter-writing skill that turns a single topic into a send-ready issue: A/B subject lines, a matching preview text, a scannable structure, the full draft in your brand voice, and a clear, benefit-driven CTA — plus a best-practice self-check for mobile, deliverability and GDPR. Real reader value, no clichés.
Website audit across 7 categories scored 0–100 (Web Vitals, security, SEO, accessibility, best practices, mobile, privacy) — one concrete fix per finding.
Web Vitals is an AI web-performance audit skill that audits a website like the established pro tools — across seven categories each scored 0–100: performance (Core Web Vitals LCP/INP/CLS + lab metrics), accessibility (WCAG 2.2), best practices, SEO, security (TLS + security headers in the style of Mozilla Observatory/SSL Labs), mobile/responsive and privacy/GDPR. Delivers severity-ranked findings with concrete fixes per category, an overall overview and a leverage-sorted action list — and honestly distinguishes measured from estimated values.
Do ChatGPT, Claude & Perplexity read and cite your site? An AI-visibility audit across 7 categories scored 0–100 — the real levers first: crawl access, raw-HTML readability, structured data and GEO. One concrete fix per finding, with a recommendation strength.
Web AI Vitals is an AI-visibility (GEO) audit skill that checks how well a website can be crawled, read, understood and cited by AI answer engines (ChatGPT, Claude, Perplexity, Google AI Overviews/Gemini) — an AI-visibility/GEO audit across seven categories scored 0–100. The real levers first: can the AI bot even get in (robots.txt, CDN/Cloudflare blocks)? Is the content in the raw HTML (AI bots mostly DON’T render JavaScript)? Are there structured data (Schema.org/JSON-LD) and citable direct answers (GEO)? Plus trust & freshness (E-E-A-T, Bing index). AI guide files like llms.txt/ai.txt come last — honestly framed as a low-leverage bonus. Gives a concrete fix per finding with a recommendation strength and a leverage-prioritized action list.
Distinctive front-end UI design that looks beautiful AND is verifiably correct — design first, accessibility and performance checks run alongside.
Frontend Design is an AI skill for front-end UI design that doesn't look AI-generated: it builds distinctive, production-ready interfaces, checked for accessibility (WCAG), contrast, focus, layout shift and performance.
Brainstorming for anything — generate ideas the guided way: clarify the goal first, then diverge broadly and converge critically to a shortlist and next steps.
Brainstorm is an AI brainstorming skill that runs a structured brainstorming session on any topic — generating ideas for products, projects, to-dos, campaigns, names, problems and decisions. Guides you question by question, produces many ideas with proven brainstorming techniques, and converges critically to a reasoned shortlist.
Turn Google Analytics data into decisions — advises your tracking setup, analyzes pasted GA4 reports anywhere, and pulls live GA4 data in code-capable tools. With comparison tables, anomaly detection and prioritized recommendations.
Analytics is an AI analytics skill that analyzes your Google Analytics 4 data and turns web, product and marketing numbers into clear actions. Three modes: advise tracking setup, analyze pasted GA4 exports/CSV/screenshots anywhere, or pull live GA4 data in code tools. Read-only, no PII.
A smart, in-character companion for your tabletop RPG (D&D, Pathfinder & more) — remembers your character across every session, recaps the story and plays them in-character with you. Never the GM, never rolls, never decides.
Pen & Paper Companion is a tabletop-RPG companion skill that keeps ONE character across every session and campaign — story, allies, enemies, spells and gear live in files you own. Say "Open Vaelis" and your AI knows it all and plays them in-character. D&D, Pathfinder, Call of Cthulhu and more, one-shot to years-long campaign.
Build your own Discord bot — from idea to production-ready. It plans, builds, reviews and extends modern bots with you: slash commands, buttons, modern modals, Components V2, GDPR and integrations like YouTube/Twitch. Architecture and safety included — not just code snippets.
Discord Bot Architect is a Discord-bot building skill — your AI companion for building your own Discord bots, from your first bot to a large community system. Instead of loose snippets you get real architecture decisions (bot or webhook?), clean slash commands, buttons, modern modals and Components V2, plus permissions, rate-limit and timeout protection, GDPR and integrations (YouTube/Twitch/RSS). It understands Discord as a complete product, UI and API system — from setup to deployment.
Turn a rough idea into an AI-ready project plan and AGENTS.md handoff — for software AND everything else.
AI Blueprint Assistant is an AI project-planning skill that turns a rough idea into a complete, AI-ready blueprint before you build — project type, flagged assumptions, scope & non-goals, structure, roadmap, risks, Blueprint Score and a copy-paste handoff (AGENTS.md / CLAUDE.md) for ChatGPT, Claude, Cursor, Codex, Copilot or Gemini. Not software-only.
Deep research that doesn't stop at the third hit: search broad, hunt primary sources, cross-check, show contradictions, cite cleanly. Neutral — it reports sources and how strong they are, never a true/false verdict.
Deep Research is an AI research skill for deep, multi-source research and fact-checking — it hunts primary sources and cross-checks them. Breaks your question into checkable claims, verifies each against independent sources, surfaces contradictions, and delivers a traceable, cited report — neutral, never a true/false verdict.
Fact-check any AI output — answers, plans, prompts, text — for hallucinations, gaps and risks, and get the better version back.
AI Quality Check is an AI-output quality-check skill: it fact-checks and reviews AI-generated output — ChatGPT/Claude answers, project plans, prompts, text, summaries or code — for hallucinations, gaps, risks and real usability, then makes it better. Eval, not vibes: claim ledger, hallucination & AI-fluff check, score + traffic light, prioritized fixes and the next best prompt. Nine modes incl. fact-check (SIFT), red-team / agent security (OWASP LLM) and a leftover scan.
Spot AI patterns in text — fairly and transparently, without selling false certainty.
AI Signal Check is an AI-writing-pattern check skill: it analyzes text for AI-typical language, structure and style patterns and explains, fairly, *why* it reads artificial, generic or too smooth. Not an "AI detector": no authorship verdict, no percentage. Instead an assessment category, signal strength 1–10, confidence, false-positive risk, flagged passages with an alternative explanation, ranked origin hypotheses and an editorial recommendation — for website, cover letter, LinkedIn, support and editorial use.
Check first. Then label clearly.
AI Transparency is an AI skill for AI content labelling, disclosure and machine marking that, for a given AI content, system, chatbot, campaign or platform post, works out whether a labelling duty can apply, **who** is responsible, **which** current rule governs it, whether the platform already offers a label — and whether that is enough. It covers the EU AI Act (Art. 50), DACH, international modules, platform functions and C2PA. Built responsible: source-first & verify-live, never "legally safe/guaranteed", not legal advice.
So AI sounds like your brand — not like AI.
Brand Voice Guardian is an AI skill that learns, documents, applies, reviews and protects your brand's voice — so AI content sounds like your brand, not like generic AI. It builds a persistent brand-voice.profile.md, reviews new content for brand fit, tone, terminology, channel and AI-generic risk, and rewrites copy on-brand. Not a rewrite tool: a brand-governance layer for AI workflows.
From a mood to a usable color decision — colors that go together, with clear roles and a real mockup.
Color Matchmaker is an AI skill for color palettes and color schemes — it finds colors that work together, gives each one a clear role (background, text, button, accent, status), checks contrast and accessibility deterministically, and shows the palette in a real mockup. It advises people with no color knowledge in plain language — not a random generator, not a logo generator.
Stress-test your decision before you commit — a structured panel of experts, stakeholders and critical counter-voices that argue it out, so you get a real second opinion instead of one answer.
Decision Board is a decision-making skill that runs a structured panel of experts, stakeholders and critical counter-voices. Instead of one AI answer you get a real second opinion: independent first judgments, a genuine debate, an assumption and risk check, a minority report and a clear, traceable recommendation.
So Google and AIs understand your website unambiguously — real entities, a clean knowledge graph, zero schema spam.
Digital Identity Architect is an AI skill for a website's digital identity — structured data, Schema.org/JSON-LD and the knowledge graph. It first understands a site's real entities, models an entity graph with stable @ids, then implements the markup framework-correctly and validates it — honestly, with no rich-result hunting. Not about digital identity in the eID/login sense.
Find the right official source, form, office and next step.
Government Navigator is an international public-services & life-admin navigation skill: it turns a life event ("I'm moving abroad", "I need a visa", "I got an official letter") into a clear, ordered route — the responsible authority, the official source, the forms, the documents, the appointment, the deadlines and the next step. It covers visas/immigration, travel authorizations (ETIAS/EES/ESTA/eTA/UK-ETA), relocation, work/study, family, benefits, business setup, apostille/legalisation, official letters and government accounts — worldwide. Built to be responsible, not confidently wrong: official-source-first, never invents fees/deadlines/URLs (verify-live), no guarantee, mandatory scam layer. Not legal advice.
Every session deserves a Happy Ending — the clean close for your AI session, without losing the thread.
Happy Ending is an AI skill for closing AI work sessions cleanly, turning the chat into the valid project state: final decisions, open tasks, a memory patch and a copy-paste next-session prompt. Its core is a Memory Safety layer, so discarded, canceled or superseded work is not mistakenly saved as the valid state.
Make AI text credible, concrete and true to your real voice — human through quality, not disguise.
Humanizer Pro is an AI skill for reworking generic or AI-typical text that makes it more natural, concrete, credible and channel-fit — without distorting facts and without "beating" detectors. It spots clichés, buzzwords and over-smoothed AI structure, protects names, numbers and core claims (Meaning Lock), and rewrites for email, website, LinkedIn, cover letters, support and more — in your real voice or your brand voice.
Find real problems instead of guessing ideas.
Pain Miner is a voice-of-customer skill for pain mining: it finds the real problems, wishes, objections, buying motives and switch/churn reasons behind a product, industry, audience, brand, competitor or your own customer data — and turns them into product, marketing, sales and content decisions. Method-grounded (Jobs-to-be-Done, The Mom Test, Value Proposition Canvas, Kano, RICE), it preserves the customer's real language (verbatim, verified), labels every finding by confidence (backed/plausible/hypothesis/weak), and prioritizes instead of listing. Honest by design: hypothesis ≠ evidence, no vocal minority as truth, no fear-mongering.
AI copywriting that turns the pain point into the punchline — and the punchline into the product benefit.
AI copywriting for marketing hooks, ad copy and campaigns: Pain-to-Punchline turns real problems and everyday frustrations into scroll-stopping hooks, headlines, punchlines and taglines — with method, not random jokes, and always landing on the product benefit.
Checks your project for data-protection risks, open questions and expert-review needs — before it goes live.
Privacy Check is an AI skill for a structured data-protection pre-check of any project — it makes visible which data, purposes, jurisdictions, vendors, consents, retention periods and risks must be checked before you go live. Not just websites: apps, newsletters, giveaways, events, leads/CRM, communities, chatbots, AI agents, shops, courses, HR and automations too. It detects the project type and use cases, routes the relevant regimes (GDPR, ePrivacy, EU AI Act, UK, US state laws), flags high-risk hard, and delivers a risk traffic light, prioritized to-dos and an expert-review briefing. Not legal advice, not a final sign-off.
Reply sober when you're not — clear thinking is enough, neat typing is optional.
Sober Reply is an AI skill for clear, safe replies in hard-to-write moments — a low-input, plain-text and safety mode that understands messy input, typos and half-formed thoughts when you're drunk, tired, stressed, sick or simply not in writing shape. It keeps replies short, asks only light questions, de-escalates messages and pushes risky actions like sending, posting, buying or deploying to a sober moment. With crisis & emergency protection and one firm rule: messy input never automatically means "drunk".
Make every shared link to your site look great everywhere — strategy first, on-brand, technically correct.
Social Preview Architect is an AI skill for Open Graph and social-share previews that plans, designs, checks and maintains the images and meta tags your website shows when links are shared on Discord, Slack, LinkedIn, WhatsApp, X and beyond. It never generates blindly: it first inspects the real page (logo, colors, theme, existing og: tags), then proposes a preview strategy for approval.
Burn your tokens — on maximum fun.
Token Waster is an AI skill that lets you waste your tokens on a few small games, right in your AI chat. Quick minigames and a few longer save-driven ones, sharing one save system, an on-demand dashboard, a story-memory engine and a cross-game profile. The joke is anti-efficiency: spend your tokens on fun instead of saving them. Everything is single-player; to play with friends you swap a challenge code. Built honest: it promises only what works everywhere, and everything better is a bonus.
Website audit for SEO, security, Core Web Vitals & AI visibility — measured against TODAY's official standards, with safe fixes.
Website audit tool for SEO, security headers, Core Web Vitals, GDPR and AI visibility (GEO). Measures your site or local project for real and checks every rule live against the current official source — then delivers a prioritized, evidence-backed report and safe fixes.
Detects the wiki engine first, then writes the right markup — instead of blindly guessing MediaWiki or Markdown.
Wiki Editor is an AI wiki-editing skill that edits pages with the correct syntax (wikitext, Markdown, Confluence): it detects the wiki engine, understands the markup, preserves local templates and infoboxes, suggests internal links, plans images, builds tables and categories, and writes the edit summary.
Skills & agents are ready-made abilities for your AI: polished, reusable building blocks for one specific task. You pick one, install it with a single command (or download it) and your AI tool does that task noticeably better right away. The platform tabs above show exactly how.
Browse by category or search and pick the skill or agent that solves exactly your task.
One command (npx) drops the skill straight into your tool — or download the ready-made folder in the open standard (SKILL.md, or a Claude Code plugin for agents). Pay once, no subscription.
Load it into your AI tool (see the platform tab) — your AI can do the new task instantly.
Thanks to the open agentskills.io standard the same skill works across many tools — Claude, ChatGPT/Codex, Cursor, GitHub Copilot, Gemini CLI and more. Agents run in Claude Code. Purchased skills install with a single command (npx) — see “My Skills”.
Claude supports skills natively ("Agent Skills", generally available). The skill activates automatically as soon as your request matches its description.
In claude.ai under Settings → Capabilities, turn on "Skills" and code execution (on a paid plan: Pro, Max, Team or Enterprise).
Under Skills, choose "Upload a skill" and upload the skill folder as a ZIP (folder name = skill name, contains the SKILL.md), then toggle it on.
Claude invokes the skill automatically when it fits. In Claude Code, instead drop the folder into ~/.claude/skills/ and start it with /skill-name.
In a Claude Project without the Skills feature: paste the SKILL.md text into the project instructions and upload reference files as knowledge.
ChatGPT does not (yet) read the SKILL.md format automatically — you transfer its content once into a custom GPT or a project. After that it behaves like the skill.
In ChatGPT click "+ Create" → GPT Builder → "Configure" tab (requires Plus, Pro, Team or Enterprise).
Paste the contents of the SKILL.md into "Instructions"; upload templates/reference files as "Knowledge" (up to 20 files).
Enable Code Interpreter if needed, then save. A lighter route is a ChatGPT Project or the "Custom Instructions".
Note: OpenAI does not allow "GPT" in your own product names — you only use the skill inside your own GPT.
The universal route for any chat model — no special feature required.
Open the SKILL.md and copy its contents (the instructions).
Paste it as a system prompt or first message; attach reference data if needed.
Write your task — the model follows the skill’s instructions.
Tools like Cursor, GitHub Copilot, Codex, Windsurf and Cline read the same standard — easiest via the npx command (see “My Skills”), which drops the files in tool-natively. Gemini CLI installs as an extension.
Claude is a trademark of Anthropic, PBC. ChatGPT and OpenAI are trademarks of OpenAI. skills-for-ai is not affiliated with or endorsed by either company.
A skill is a ready-made instruction file (SKILL.md in the open agentskills.io standard) that lets your AI tool handle a specific task reliably. Pick it, download it, drop it into your tool — done.
A skill is a portable instruction file that runs in almost any AI tool. An agent is a ready-to-run plugin for Claude Code (and compatible tools): it bundles a command, an expert persona and several skills, fetches official sources live when needed, and works locally with your own Claude. In the library, agents are labeled “Agent”; everything else is a portable skill.
Your account is created automatically at checkout — just your email. Payment runs securely through Lemon Squeezy as the seller of record: one-time purchase, no subscription. Afterwards everything stays available under “My Skills & Agents” for re-download at any time, including your license key.
Download the file and add it to your tool: in Claude as a Project/system prompt, in ChatGPT as a Custom GPT or custom instruction, in Cursor/Copilot/Gemini CLI as a rule. Nothing to set up beyond pasting or uploading the file.
No. Skills are plain instruction files — you add them to your AI tool and work in normal language. There are developer skills too, but using a skill needs no code.
One-time. You download a file and keep it — no subscription, no recurring fee.
Any tool that reads the open agentskills.io standard — Claude, ChatGPT/Codex, Cursor, GitHub Copilot, Gemini CLI and more. Each skill lists its compatible tools.
A SKILL.md file (plus any extra files a skill needs) in the open agentskills.io format — human-readable and usable across many AI tools.
A prompt is a single instruction you type each time. A skill is a structured, reusable instruction file: add it once and the AI keeps the same reliable context and rules for the whole task — no retyping.
Usually yes. You can paste a skill’s instructions into a normal chat or set them as a custom instruction. Convenience features like Projects or Custom GPTs may need a paid plan, but the skill itself is plain text you can use anywhere.
Yes. Because skills follow the open agentskills.io standard, the same file works in Claude, ChatGPT, Cursor, Copilot, Gemini CLI and more — get it once, use it everywhere.
Yes. A skill is a plain text file (SKILL.md) — open it and adapt the instructions to your needs, your brand voice or your workflow.
A skill is just an instruction file — it collects no data and sends nothing anywhere. Your inputs stay in whatever AI tool you use, under that tool’s privacy terms.
agentskills.io is an open format for AI skills: a SKILL.md file with a short description and instructions. Because it’s open, one skill works across many AI tools instead of being locked to one.
Passwordless — magic link or social login.
Check your inbox — we sent you a login link.
Something went wrong. Please try again.
