# ToolGenX — Full Documentation & Catalog > Last-Updated: 2026-09-21 > Comprehensive catalog and raw text for ToolGenX digital products, Kindle books, free calculators, and technical blog posts. Built for LLMs, AI search engines (ChatGPT, Claude, Perplexity, Gemini, Google AI Overviews), and agents. ## About ToolGenX - **Website:** https://www.agentskillpacks.com - **Founder:** İsmail Günaydın (Software Engineer & SEO/GEO/AEO Strategist) - **Legal Owner:** Melike Günaydın (Software Engineer; Turkish Sole Proprietorship registered in Ümraniye / İstanbul, Tax ID: 1470465320, Alemdağ Tax Office) - **Contact:** support@toolgenx.com · +90 506 545 88 35 · KEP: melike.gunaydin@hs01.kep.tr - **Model:** Independent digital product shop. One-time payment, instant download, zero subscription. - **Refund Policy:** 14-day refund for undownloaded products per EU CRD Art 16(m). 30-day replacement for defective files. ## 1. Digital Products Catalog (19 Items) Every product is a one-time USD purchase with instant delivery. ### AI Agency Launch Kit ($119 USD) - **URL:** https://www.agentskillpacks.com/products/ai-agency-launch-kit - **Category:** Business & Growth - **Type:** toolkit - **Badge:** Bestseller - **Tagline:** Build a running AI agency with a real system, not guesswork. - **Summary:** A 37-skill system that walks through every stage of starting an AI agency — finding clients, delivering work, and getting paid on retainer. - **Target Audience:** Freelancers and consultants building an AI-powered agency - **Pain Point Solved:** You want to run an AI agency but you are stitching things together from YouTube videos and hoping for the best. - **What Ships in the Box:** 37 battle-tested skills, a 6-phase workflow, pricing templates, cold email sequences, and a full client delivery system. #### Key Features - 37 Claude Code skills covering the full agency workflow - 6-phase system: prospect, pitch, deliver, retain, upsell, scale - Done-for-you cold email, proposal, and onboarding templates - GEO/SEO service packages priced and ready to sell - Lead research and outreach automation included #### Frequently Asked Questions - **Q: How long until I land my first client?** **A:** Most buyers send their first cold email within a week of unzipping the kit. First paid retainer typically lands inside 30-60 days if you actually run the prospect phase daily. The kit removes the planning excuse, not the doing. - **Q: Do I need to use Claude Code for this?** **A:** Yes, the 37 skills are Claude Code skills that drop into your `~/.claude/skills` folder. The cold email scripts, proposals, and pricing templates are plain Markdown and work in any tool. About 60% of the value runs with or without Claude Code. - **Q: What if I am brand new to agency work?** **A:** The 6-phase system assumes you know nothing about running an agency. It starts at picking a niche and ends at retaining a $5K-$10K monthly client. There is no "you should already know" gap baked in. - **Q: Can I use it for SEO, GEO, or general AI consulting?** **A:** All three. The skill set covers GEO/SEO audits, AI integration consulting, content systems, and growth automation. The pricing templates include packaged offers for each so you can sell whichever fits your background. - **Q: Lifetime updates included?** **A:** Yes — every v1.x update lands in your account at no extra cost. Major future versions (v2) ship as a discounted upgrade for existing buyers. The agency landscape moves fast, so I keep this one current. --- ### AI CFO Toolkit ($29 USD) - **URL:** https://www.agentskillpacks.com/products/ai-cfo-toolkit - **Category:** Business & Growth - **Type:** toolkit - **Badge:** Hot - **Tagline:** Make money decisions like a CFO, even if you have never had one. - **Summary:** A financial intelligence toolkit that gives founders real visibility into runway, unit economics, pricing strategy, and scenario modeling. - **Target Audience:** Founders and operators who need financial clarity without hiring a CFO - **Pain Point Solved:** You are making big money decisions on gut feel because you cannot afford a CFO and spreadsheets are not cutting it. - **What Ships in the Box:** Financial modeling tools, pricing frameworks, scenario planning, and a CFO briefing template — one toolkit. #### Key Features - Runway calculator and cash burn tracker - Unit economics breakdown (CAC, LTV, payback period) - Van Westendorp pricing framework - Scenario modeler with conservative, base, and optimistic projections - CFO-style briefing doc for investors #### Frequently Asked Questions - **Q: Do I need accounting knowledge to use this?** **A:** No. The toolkit translates CFO concepts (runway, LTV, payback period) into plain language with worked examples using small-startup numbers. If you can fill in a spreadsheet, you can use every piece. - **Q: What spreadsheet format does it ship in?** **A:** Google Sheets and Excel-compatible .xlsx, both included. The Van Westendorp pricing tool uses sheet-native formulas, no add-ons or scripts. Models open in Numbers on macOS too, though one chart loses formatting. - **Q: Is this overkill for a pre-revenue startup?** **A:** Not in my experience. The runway calculator and Van Westendorp pricing matter most before you launch. Unit economics modeling becomes useful the week after your first sale, and the investor brief unlocks once you have any traction story. - **Q: How is this different from just hiring a fractional CFO?** **A:** A fractional CFO runs $1,500-$3,000 a month and still needs you to gather the data. This toolkit gives you the same models a CFO would build, plus a structured way to present them to investors. Use both if you can afford it. - **Q: Can I use it for client work?** **A:** Yes, the single-user license covers your own consulting engagements. You can run the models for clients and present findings. Reselling the spreadsheets themselves is not allowed. --- ### AI Content Blueprint ($29 USD) - **URL:** https://www.agentskillpacks.com/products/ai-content-blueprint - **Category:** Content & Writing - **Type:** toolkit - **Tagline:** Write articles that rank on Google AND get cited by ChatGPT. - **Summary:** A 14-section article framework built for AI search. Covers structure, quick-answer boxes, FAQ schema, and how to write content that gets pulled into AI answers. - **Target Audience:** Bloggers, content marketers, and SEO writers who want both Google and AI search traffic - **Pain Point Solved:** You spend 5 hours writing a blog post, it gets 12 views, and ChatGPT has never mentioned your site once. - **What Ships in the Box:** 14-section content framework, quick-answer box templates, FAQ schema blueprints, and a blog writer skill for Claude Code. #### Key Features - 14-section article framework for Google and AI search - Quick answer box formula that gets picked up by ChatGPT and Perplexity - FAQ schema setup that triggers rich results - Blog writer skill for Claude Code, fully automated - Step-by-step guide to writing content AI engines actually cite #### Frequently Asked Questions - **Q: Will this work without Claude Code?** **A:** The 14-section framework, quick-answer formulas, and FAQ schema templates are tool-agnostic — they work in any editor. The blog writer automation skill needs Claude Code. About 70% of the kit delivers value with or without it. - **Q: How fast do AI citations start showing up?** **A:** In my own posts, Perplexity started citing inside two weeks and ChatGPT picked up content around the four-to-six week mark after the schema and structure changes landed. Sites with stronger existing authority see it faster; new domains take longer. - **Q: Is the framework just for blog posts?** **A:** No. The 14 sections map to long-form articles, but the quick-answer + FAQ + schema pieces apply to product pages, documentation, and landing copy. I use the same shape on every page that wants AI citations. - **Q: Does it replace my SEO process?** **A:** It extends it. Traditional SEO (keywords, intent, links) still matters for Google. The blueprint layers GEO-specific structure on top so the same content earns both rankings and AI answer-box mentions. - **Q: How long is the implementation curve?** **A:** Reading the framework takes about 90 minutes. Restructuring an existing post against it takes around an hour. Writing a new post inside the framework, end to end, sits at two to three hours once you have done two or three. --- ### AI Content Empire Builder ($29 USD) - **URL:** https://www.agentskillpacks.com/products/ai-content-empire-builder - **Category:** Content & Writing - **Type:** toolkit - **Badge:** Popular - **Tagline:** One post becomes 20 pieces of content. Stop creating from scratch every time. - **Summary:** An 8-skill content multiplication system that turns one piece of content into social posts, threads, newsletters, and more — without writing everything separately. - **Target Audience:** Content creators and solo founders who need to show up consistently but are running out of time - **Pain Point Solved:** You are posting every day but it takes 3 hours per post and your numbers still do not move. - **What Ships in the Box:** 8 Claude Code skills, a content repurposing calendar, AI humanizer, and platform-specific optimizers. #### Key Features - 8 production-ready skills for content creation and repurposing - Content calendar system: one piece becomes a week of posts - AI humanizer that removes robotic tone from AI writing - Twitter/X algorithm optimizer for more reach with less effort - Works with Claude Code, no extra subscriptions needed #### Frequently Asked Questions - **Q: How many platforms can I cover with one source post?** **A:** The calendar maps one anchor post into 12 deliverables: 3 LinkedIn posts, 2 Twitter/X threads, 1 newsletter, 1 short-form video script, 3 carousel slides, and 2 reply hooks. Reasonable single person can ship the full set across a week. - **Q: Will the output sound robotic?** **A:** The AI humanizer skill is the third thing the pipeline runs. It strips 50 detection-triggering patterns and matches voice from samples you provide. You still want to skim the output, but I have not had a client flag a piece as AI-written in six months. - **Q: Do I need to write the source post myself?** **A:** Yes — the system multiplies content, it does not invent your point of view. Bring a 600-1500 word post or transcript and the skills handle the formatting and adaptation. Garbage in, polished garbage out. - **Q: Does it cover short-form video?** **A:** Scripts and hook variations, yes. Actual video generation is not in this pack — pair it with ShortFactory AI for end-to-end YouTube Shorts. The two are designed to share the same source post. - **Q: How long does the first run take?** **A:** About 45 minutes once the skills are installed. You set the source post, walk through the 8 skills in order, and end with a packed week of content. Subsequent runs settle around 25 minutes with practice. --- ### AI Search Visibility Toolkit ($49 USD) - **URL:** https://www.agentskillpacks.com/products/ai-search-visibility-toolkit - **Category:** SEO & Visibility - **Type:** toolkit - **Badge:** Hot - **Tagline:** If ChatGPT and Perplexity do not mention you, you are invisible to half the internet. - **Summary:** An 11-skill GEO (Generative Engine Optimization) system that audits, scores, and fixes your site so AI search engines find, trust, and cite your content. - **Target Audience:** SEO professionals, agency owners, and business owners who want AI search traffic - **Pain Point Solved:** Google is sending AI Overviews now. Perplexity is citing your competitors. You are not on the list. - **What Ships in the Box:** 11 GEO audit skills, an llms.txt template, citation scoring system, and a client-ready PDF report generator. #### Key Features - Full GEO audit scoring your site across 11 AI visibility factors - llms.txt setup so AI crawlers can read your site - Client-ready PDF report generator for agencies - Citation audit — see where AI engines stand on your content - Works with ChatGPT, Perplexity, Google AI Overviews, and Gemini #### Frequently Asked Questions - **Q: How long before ChatGPT or Perplexity start citing my site?** **A:** Perplexity tends to react fastest, usually within 7-14 days of structural fixes if your domain has any existing authority. ChatGPT and Google AI Overviews are slower — 4-8 weeks is a realistic window. Brand-new domains take longer everywhere. - **Q: Is this for agencies or for my own site?** **A:** Both. The PDF report generator targets agency client deliverables and produces white-label output. The same 11 audit skills work on your own site without any change. Most buyers run both modes. - **Q: Do I need to be technical to use it?** **A:** Some pieces require editing robots.txt and uploading an llms.txt file, which is one git commit or one FTP upload. The audit and scoring skills run inside Claude Code without any code. If you can deploy a one-line file, you can use everything. - **Q: Will it flag issues that other SEO tools miss?** **A:** Yes — most SEO tools score for Google rankings, not for AI citation readiness. The 11 GEO factors here cover passage-level citability, llms.txt presence, structured data depth, and bot accessibility — none of which Ahrefs or Semrush report on. - **Q: Lifetime updates?** **A:** Yes, including new AI search engines as they ship. When OpenAI Atlas or Gemini Deep Research land citations, the audit grows to cover them. v1.x updates are free for buyers. --- ### AI SEO Command Suite ($69 USD) - **URL:** https://www.agentskillpacks.com/products/ai-seo-command-suite - **Category:** SEO & Visibility - **Type:** toolkit - **Badge:** Bestseller - **Tagline:** 30 SEO skills in one place. Run audits, find keywords, fix your site from the terminal. - **Summary:** A 30-skill SEO command system for Claude Code covering keyword research, competitor analysis, content gaps, schema generation, and GEO optimization in one toolkit. - **Target Audience:** SEO professionals, agency owners, and developers who want to automate SEO work - **Pain Point Solved:** You are juggling 8 different SEO tools, spending $600 a month on subscriptions, and still doing half the work manually. - **What Ships in the Box:** 30 production-ready SEO skills, keyword research templates, competitor analysis workflows, and a full schema library. #### Key Features - 30 Claude Code skills covering the full SEO workflow - Keyword research, SERP analysis, and content gap finder - Competitor audit showing what they rank for that you do not - Schema markup templates for rich results and AI citation - GEO and traditional SEO combined into one command system #### Frequently Asked Questions - **Q: Can this replace my Ahrefs or Semrush subscription?** **A:** Partly. The skills cover the workflow side — research, analysis, content planning, schema. For live ranking data and backlink indexes you still want a paid tool, but you can downgrade to a cheaper tier because the heavy lifting moves into Claude Code. - **Q: Do I need API keys for SERP data?** **A:** Optional. Several skills use DataForSEO or SerpApi if you have a key; the rest use Claude's web tool. The pack includes a free-tier playbook that gets you 80% of the value without paid APIs. - **Q: Is this Google-only or does it cover AI search too?** **A:** Both. Traditional Google SEO is the bulk, but the suite includes GEO-specific skills for AI Overviews, ChatGPT citations, and Perplexity. If you want AI search depth specifically, pair this with the AI Search Visibility Toolkit. - **Q: How is this different from your free SEO content online?** **A:** The free posts cover concepts. This pack is the executable form — 30 Claude Code skills you run from the terminal, plus templates and a schema library. Honest answer: read the blog first, buy this when the manual work starts hurting. - **Q: Will it work for non-English sites?** **A:** Yes, the skills are language-agnostic. The schema library, keyword templates, and audit logic all work in Turkish, Spanish, German, French and Arabic. Voice-tuning prompts may need light translation work for non-Latin scripts. --- ### AutoEarnaAI ($27 USD) - **URL:** https://www.agentskillpacks.com/products/autoearnaai - **Category:** Automation - **Type:** toolkit - **Badge:** Popular - **Tagline:** Four AI automation tools that run on your computer — no cloud fees, no monthly bill. - **Summary:** A local AI toolkit with four automation tools: YouTube Shorts creator, Twitter/X content bot, Amazon affiliate finder, and local business outreach engine. Powered by Ollama, zero API cost. - **Target Audience:** Side hustlers, affiliate marketers, and online income seekers who want automation without subscriptions - **Pain Point Solved:** You want to make money online but you are manually doing everything that could run on autopilot. - **What Ships in the Box:** 4 Python automation tools, local AI setup guide, no-subscription model, and step-by-step launch instructions. #### Key Features - YouTube Shorts automation: script, voiceover, captions, upload - Twitter/X content bot that posts while you sleep - Amazon affiliate product finder that auto-generates review content - Local business outreach engine: find leads, send cold emails - Runs on Ollama locally — one-time payment, zero ongoing API costs #### Frequently Asked Questions - **Q: Will this actually make me money?** **A:** It removes the manual labor, which is the part that kills most side hustles before they earn anything. Whether you make money depends on the niche you pick and how long you let the automation run. I have seen affiliate sites hit $300-$1,200 a month after 90 days of daily posting. - **Q: Do I need to know Python?** **A:** No. The four tools ship as ready-to-run scripts with a setup checklist. You install Ollama, copy two config files, and run the launchers. Editing the prompts is the only place you might touch a config — and that is a text file, not code. - **Q: How much does it cost to run after the one-time purchase?** **A:** Zero recurring API fees by default — Ollama runs locally on your machine. Optional ElevenLabs voice or premium model upgrades add usage costs only if you opt in. Most buyers run everything on free local models for the first 90 days. - **Q: What hardware do I need?** **A:** A machine with 16 GB RAM handles the standard local models smoothly. 8 GB works with smaller models and slower output. M-series Macs, modern PCs, or any GPU-equipped machine all work. There is a Raspberry Pi mode for the Twitter bot only. - **Q: Are these tools allowed on YouTube, Twitter, Amazon?** **A:** Yes within current Terms of Service if you do not spam — the automation generates content, you still review and publish. Mass-posting unedited output gets accounts flagged. The guide spells out what trips each platform. --- ### B2B Growth Machine ($49 USD) - **URL:** https://www.agentskillpacks.com/products/b2b-growth-machine - **Category:** Business & Growth - **Type:** toolkit - **Badge:** Hot - **Tagline:** Seven systems that fill your pipeline, close deals faster, and track revenue properly. - **Summary:** A B2B growth stack with 7 interconnected systems: outbound engine, ICP builder, cold email sequences, content-to-pipeline attribution, deal scaling playbook, and revenue intelligence. - **Target Audience:** B2B founders, sales teams, and growth leads who need a repeatable pipeline system - **Pain Point Solved:** Your pipeline is guesswork. Cold emails get ignored. You close one deal and have no idea how to repeat it. - **What Ships in the Box:** 7 growth systems, ICP templates, cold email sequences, a deal scaling playbook, and revenue tracking tools. #### Key Features - ICP builder — define who you should actually be targeting - Cold email sequences that get replies, not spam-folder deliveries - Prospect research automation to find the right people fast - Deal scaling playbook from $1K to $100K contracts - Revenue intelligence dashboard for tracking what works #### Frequently Asked Questions - **Q: Is this only for cold outbound?** **A:** No. Two of the seven systems cover inbound and content-to-pipeline attribution. The cold email piece is the loudest, but the ICP builder and deal scaling playbook apply to any GTM motion — including PLG and partner-led. - **Q: What deal size does this fit best?** **A:** Built around the $1K-$100K annual contract range. The scaling playbook walks through the path from $1K pilots to $20K-$100K annual. Sub-$500 monthly self-serve does not need this depth. - **Q: Do I need a CRM already set up?** **A:** A CRM helps but is not required for day one. The revenue intelligence dashboard ships as a Sheets template with HubSpot, Pipedrive, and Attio import patterns documented. Switch from sheet to CRM when monthly deal volume crosses 15. - **Q: How long until pipeline starts moving?** **A:** First responses on cold sequences land in 5-10 days if your ICP and offer are sharp. First qualified meeting in 2-4 weeks for sub-$10K deals. The playbook is honest about the ramp curve — overnight pipeline is not on the table. - **Q: Will the cold emails go to spam?** **A:** Not if you warm the sending domain first. The pack includes a 14-day warmup checklist (SPF/DKIM/DMARC, gradual volume, reputation tools). Skip warmup and you spam-fold yourself — no growth system fixes that. --- ### CodeForge ($19 USD) - **URL:** https://www.agentskillpacks.com/products/codeforge - **Category:** Developer Tools - **Type:** template - **Badge:** Popular - **Tagline:** Stop asking AI to write code. Start letting it run your entire dev workflow. - **Summary:** A 14-skill AI coding workflow system for Claude Code covering planning, TDD, code review, debugging, git worktrees, and multi-agent development across Cursor, Codex, and Gemini CLI. - **Target Audience:** Developers who use AI coding tools but feel they are not getting the most out of them - **Pain Point Solved:** You use Cursor or Claude to write code but your workflow is still slow, messy, and full of back-and-forth fixes. - **What Ships in the Box:** 14 Claude Code skills, a complete TDD workflow, planning system, code review automation, and git worktree setup. #### Key Features - 14 dev workflow skills for Claude Code - TDD workflow: red-green-refactor with subagent support - Planning and brainstorming system before you write a single line - Git worktrees integration for parallel development branches - Works with Claude Code, Cursor, Codex, and Gemini CLI #### Frequently Asked Questions - **Q: Does it work with Cursor, Codex, or Gemini CLI?** **A:** Yes, the skills include adapter configs for Cursor (custom-instructions), Codex CLI, and Gemini CLI. Claude Code is the primary target since skills are native there. Adapters get 80-85% of the value in other tools. - **Q: I already use Cursor — what is different here?** **A:** Cursor gives you a chat with your repo. CodeForge gives you a workflow that runs that chat the same way every time: red-green-refactor TDD, parallel worktrees for two-track development, automated code review. The skills enforce the discipline you keep meaning to follow. - **Q: What languages are supported?** **A:** The workflow is language-agnostic. Examples ship in TypeScript, Python, and Go. The TDD red-green-refactor pattern, planning skill, and code review skills work identically across languages. Test runner configs cover Vitest, Jest, pytest, and Go test. - **Q: Is the git worktree integration safe?** **A:** Yes — worktrees are a native git feature with zero risk to the main checkout. The included scripts wrap `git worktree add/remove` with cleanup safeguards. The pattern lets you run parallel feature branches without stashing context every time. - **Q: Will this slow me down at first?** **A:** Yes, for about a week. TDD and the planning gate feel like extra steps. Around the third or fourth feature the speedup catches up; after that you ship faster because the rework loop shrinks. The first week is the cost. --- ### Conversion Rate Domination ($29 USD) - **URL:** https://www.agentskillpacks.com/products/conversion-rate-domination - **Category:** Business & Growth - **Type:** toolkit - **Tagline:** Getting traffic but nobody buys? Fix the funnel — one conversion at a time. - **Summary:** An 8-skill CRO suite that audits landing pages, optimizes forms and popups, sets up A/B tests, and applies marketing psychology to turn visitors into buyers. - **Target Audience:** E-commerce owners, SaaS founders, and marketers with traffic but weak conversion rates - **Pain Point Solved:** You are paying for ads, getting clicks, and watching people leave without buying anything. A leaky bucket. - **What Ships in the Box:** 8 CRO skills, landing page audit templates, A/B testing workflows, and a marketing psychology playbook. #### Key Features - 8 CRO skills covering landing pages, forms, popups, and onboarding - Landing page audit tool that finds where people drop off - A/B test setup for any element — no developer needed - Marketing psychology principles baked into every optimization - Funnel analysis from first click to checkout #### Frequently Asked Questions - **Q: Will this work on my e-commerce store?** **A:** Yes — the funnel, popup, and form skills apply directly to Shopify, WooCommerce, and BigCommerce. The marketing psychology playbook is platform-agnostic. SaaS onboarding flows have a dedicated skill in the pack as well. - **Q: How much traffic do I need before A/B testing is worth it?** **A:** Statistical significance needs around 1,000 conversions per variant per month at minimum. Below that, run qualitative tests (heatmaps, session replays, customer interviews) which the audit skill covers. The pack tells you which method fits your traffic level. - **Q: Do I need a developer to implement changes?** **A:** Most of the form and popup wins ship as no-code changes in your page builder. A/B test setup ships as Google Optimize replacement scripts (since Google Optimize sunset) and works with any analytics platform. Heavy redesigns still want a developer. - **Q: How long until I see CR uplift?** **A:** First wins from the audit usually land inside two weeks — popups, form-field reduction, headline rewrites. Compounding gains from systematic A/B testing build over 60-90 days. The playbook is conservative on timelines. - **Q: Does it cover paid traffic landing pages?** **A:** Yes, the landing page audit skill has a dedicated paid-traffic scoring path that weighs message-match, scroll depth, and bounce. Different from organic landing pages because intent is different. Both modes are covered. --- ### Humanizer Pro ($19 USD) - **URL:** https://www.agentskillpacks.com/products/humanizer-pro - **Category:** Content & Writing - **Type:** toolkit - **Badge:** Bestseller - **Tagline:** AI wrote it. Now nobody can tell. Make content sound like a real person wrote it. - **Summary:** A 3-skill content humanization system that strips AI writing patterns, removes banned phrases, matches your voice, and passes AI detection tools without making the content worse. - **Target Audience:** Writers, agency owners, and content marketers who use AI and need it to pass as human-written - **Pain Point Solved:** Your clients can tell it is AI. Detection tools flag it. And the writing sounds like a press release from 1998. - **What Ships in the Box:** 3 content skills, the 50-phrase detection list, voice matching system, and a before/after comparison workflow. #### Key Features - 50 banned AI phrases that trigger detection tools — all removed - Voice matching so output sounds like you, not a chatbot - Before/after comparison mode so you see exactly what changed - Works on blog posts, emails, LinkedIn, sales copy, and more - 3 skills: humanizer, copy editor, and copywriter in one pack #### Frequently Asked Questions - **Q: Does it actually fool AI detectors?** **A:** GPTZero and Originality.ai score humanized output at 5-15% AI probability in my tests, versus 85-99% for raw GPT-4 output. No tool is perfect — detector vendors update weekly — but Humanizer Pro stays current against the patterns that trigger flags. - **Q: Will it change my message?** **A:** No. The skills rewrite patterns, not content. Statistics, claims, names, and arguments stay intact. The before/after comparison view lets you spot any drift in a single side-by-side pass. - **Q: Can I train it on my own voice?** **A:** Yes — paste 3-5 paragraphs of your past writing into the voice profile and the humanizer biases output toward your sentence shape, word choice, and rhythm. The result is much closer to "you wrote it" than generic neutral. - **Q: How long does humanizing a post take?** **A:** About 90 seconds per 1,000 words to run, 5-10 minutes to skim the before/after diff and accept changes. Adding the voice profile step adds 2 minutes the first time, then auto-applies. - **Q: Does it work in languages other than English?** **A:** English is fully covered. Turkish, German, Spanish, and French have working profiles but with smaller training sets — output quality is acceptable but the AI-tells removal is less complete. Arabic, Japanese, Chinese are not supported in v1. --- ### Indie Game Studio Kit ($39 USD) - **URL:** https://www.agentskillpacks.com/products/indie-game-studio-kit - **Category:** Developer Tools - **Type:** toolkit - **Badge:** New - **Tagline:** One developer. 27 AI agents. A game studio running on your machine. - **Summary:** A 27-agent AI game development system that gives solo indie devs a full virtual studio covering design, programming, narrative, audio, QA, and release management for Godot, Unity, and Unreal. - **Target Audience:** Solo indie game developers who want to build bigger games without hiring a team - **Pain Point Solved:** You are a solo dev with a vision for a real game but it is just you — no designer, writer, QA, or composer. - **What Ships in the Box:** 27 specialized AI agents, multi-agent collaboration system, engine-specific specialists, and a full game dev workflow. #### Key Features - 27 specialist AI agents covering every game dev discipline - Godot, Unity, and Unreal Engine specialist agents included - Narrative director, economy designer, and audio director agents - Multi-agent workflow — agents collaborate on the same project - Built for solo devs who want studio-quality output without a studio #### Frequently Asked Questions - **Q: Which game engines does it support?** **A:** Godot 4, Unity (2022 LTS and 6.x), and Unreal Engine 5 each have dedicated specialist agents. The narrative, audio, economy, and QA agents are engine-agnostic. About 30% of the agents are universal. - **Q: I have never shipped a game — is this too advanced?** **A:** The kit assumes you can already open the editor and follow a tutorial. If you have never built a prototype, start with Godot's free first-game tutorial, then come back. The agents accelerate, they do not replace, the learning curve. - **Q: Do the agents actually write code that runs?** **A:** Yes — the engine specialist agents produce GDScript, C#, and Blueprint-friendly C++ that compiles. You still review, test, and integrate. They are not autonomous; they are pair programmers with specialist knowledge. - **Q: Can the agents collaborate on the same project?** **A:** Yes, that is the headline feature. The multi-agent workflow lets the narrative director hand off to the dialog writer, who hands off to the audio director — all working from the same project context. Coordination overhead is the cost. - **Q: Will Claude usage cost get out of hand?** **A:** For a small indie project, expect $30-$80/month in API costs if you run the agents heavily. The kit includes prompt-caching patterns and "cheap model first" routing to keep spend down. Sonnet for code, Haiku for chatter, Opus only for plot. --- ### iOS Liquid Glass UI Kit ($29 USD) - **URL:** https://www.agentskillpacks.com/products/ios-liquid-glass-ui-kit - **Category:** Design & UI - **Type:** template - **Badge:** New - **Tagline:** iOS 26 dropped. Your app looks outdated. Fix it this weekend with 67 ready styles. - **Summary:** A 5-skill design system with 67 Liquid Glass UI variants for iOS 26 covering SwiftUI, UIKit, and web adaptations with CSS backdrop filters. Build the new look without starting from scratch. - **Target Audience:** iOS developers, web developers, and UI designers who need to adopt the new iOS 26 design language - **Pain Point Solved:** iOS 26 Liquid Glass is everywhere and your app still looks like 2022. Users notice. App Store rankings notice. - **What Ships in the Box:** 67 UI style variants, SwiftUI and UIKit code, web CSS adaptations, and a design system generator. #### Key Features - 67 Liquid Glass UI style variants — ready to copy and use - SwiftUI and UIKit code for every variant - CSS backdrop filter web adaptation for cross-platform use - Design system generator: pick a style and get the full system - 5 skills covering the full iOS 26 design language #### Frequently Asked Questions - **Q: Do I need Xcode 26 to use these?** **A:** For the SwiftUI Liquid Glass effects to render exactly as designed, yes — iOS 26 / Xcode 26 is the target. The UIKit variants ship with backward-compatible fallbacks for iOS 17 and iOS 18, and the CSS web adaptation works in any modern browser. - **Q: What file format do the variants ship in?** **A:** Each variant is a self-contained Swift file (SwiftUI View or UIKit UIView subclass) plus a matching .css file for the web port. No Xcode project — drop files into your own project. Figma source files are included for the designer review pass. - **Q: Will Apple reject my app for copying the system look?** **A:** No. These variants implement the public Liquid Glass APIs Apple introduced in iOS 26. Using public APIs is exactly what they want third-party apps to do. The kit avoids private API tricks that would risk rejection. - **Q: Does the web CSS version look the same as iOS?** **A:** Close, not identical. CSS backdrop-filter and color-mix get within 90% of the native effect. The remaining 10% — animation curves, gesture responses — is where native still wins. The web port is for marketing pages and dashboards, not full app parity. - **Q: How do I pick variants without going through all 67?** **A:** The design system generator skill takes a product description and outputs the 5-7 variants that fit your app shape. The README also groups variants by use case (cards, sheets, nav, controls) so manual browsing stays quick. --- ### MarketFlow AI ($29 USD) - **URL:** https://www.agentskillpacks.com/products/marketflow-ai - **Category:** Automation - **Type:** toolkit - **Badge:** Bestseller - **Tagline:** 11 marketing systems and 42 Python scripts that actually connect your tools. - **Summary:** A marketing automation stack with 42 Python scripts that tie together HubSpot, Apollo, Gong, and Ahrefs covering A/B testing, content ops, lead scoring, cold email, and revenue intelligence. - **Target Audience:** Marketing teams and growth leads at B2B companies using HubSpot and Apollo - **Pain Point Solved:** Your marketing stack has 12 tools and they do not talk to each other. Data is everywhere. Insights are nowhere. - **What Ships in the Box:** 11 marketing systems, 42 ready-to-run Python scripts, platform integrations, and a revenue intelligence dashboard. #### Key Features - 42 Python automation scripts covering the full marketing stack - Native integrations: HubSpot, Apollo, Gong, and Ahrefs - Lead scoring automation: know who to call and when - A/B testing engine, cold email sequencer, and SEO ops tools - Revenue intelligence dashboard tying content to closed deals #### Frequently Asked Questions - **Q: Do I need API keys for HubSpot, Apollo, Gong, and Ahrefs?** **A:** Only for the integrations you use. Each script declares which keys it needs at the top. You can buy the kit, run the keyword research and content ops scripts on day one with no keys, and add HubSpot/Apollo later when you need attribution. - **Q: I am on Salesforce — can I still use this?** **A:** Partly. The HubSpot scripts have Salesforce-equivalent forks in the pack covering the same logic. Gong and Ahrefs integrations work identically. Migration time from HubSpot to Salesforce mode is about an hour. - **Q: Do I need to be a Python developer?** **A:** You need to be able to run `python script.py` from a terminal and edit a config file. That is the bar. No script writing or debugging is expected — the 42 scripts are ready-to-run, each with a one-page setup checklist. - **Q: How is this different from Zapier or Make?** **A:** Zapier is great for 1-to-1 integrations under 1,000 runs/month. MarketFlow handles multi-step pipelines, custom logic, and revenue intelligence joins that Zapier cannot price-fit. Different tools, different jobs — most teams use both. - **Q: Will the revenue intelligence dashboard work without all 4 platforms?** **A:** It works with just HubSpot and any one ad platform, but the value gets thinner. Full insight needs at least HubSpot + Ahrefs + Apollo so content, intent, and pipeline can all join. Without that, you get a fancier funnel report. --- ### Podcast to Empire Kit ($29 USD) - **URL:** https://www.agentskillpacks.com/products/podcast-to-empire-kit - **Category:** Content & Writing - **Type:** toolkit - **Badge:** Popular - **Tagline:** Record once. Show up on six platforms. Stop leaving reach on the table. - **Summary:** A 5-skill content multiplication system that turns one podcast episode into 12 content pieces — Twitter threads, LinkedIn posts, newsletters, blog posts, and short clips — automatically. - **Target Audience:** Podcasters who want more reach without recording more episodes - **Pain Point Solved:** You spend 3 hours recording and editing, then post it once on Spotify and wonder why numbers are not growing. - **What Ships in the Box:** 5 content production skills, a 12-format content system, a weekly calendar template, and a guest episode workflow. #### Key Features - One episode becomes 12 content pieces across 6 platforms - Weekly content calendar fully mapped from one recording - Guest workflow with co-promotion and cross-distribution - Twitter threads, LinkedIn posts, newsletter, blog post — auto-generated - 5 skills covering every part of podcast content operations #### Frequently Asked Questions - **Q: How long does it take to set up after I buy?** **A:** About 45 minutes to install the 5 Claude Code skills, configure your platform handles, and walk through one practice episode. After that, each new episode runs through the pipeline in 30-40 minutes for a full week of distribution. - **Q: Do I need editing software?** **A:** For the audio and short-clip outputs, yes — Descript, Riverside, or Premiere all work. The pack includes prompt patterns for each. The text outputs (LinkedIn, Twitter, blog post, newsletter) need no editing software, just a publishing tool. - **Q: Does it work for solo and interview shows?** **A:** Both. The guest workflow has dedicated patterns for interview shows: pre-show research brief, co-promotion content for the guest, and post-show recap formats. Solo shows skip those and use the standard pipeline. - **Q: How many platforms can I cover from one episode?** **A:** Twelve content pieces across six platforms: LinkedIn (3), Twitter/X (2), YouTube Shorts script (1), Instagram carousel (1), newsletter (1), blog post (1), reply hooks (3). You pick the subset that fits your audience. - **Q: Will this work for video podcasts too?** **A:** Yes. Audio-only is the primary target, but the short-clip and YouTube Shorts paths assume you have video. If you record on Riverside or Zoom with recording on, the pipeline grabs both tracks. --- ### ShortFactory AI ($29 USD) - **URL:** https://www.agentskillpacks.com/products/shortfactory-ai - **Category:** Automation - **Type:** toolkit - **Badge:** Hot - **Tagline:** Make YouTube Shorts for $0.11 each — fully automated, start to finish. - **Summary:** A Python pipeline that creates YouTube Shorts automatically: AI script writing with Gemini, visuals, ElevenLabs voiceover, auto-captions, and direct upload — at 11 cents a video. - **Target Audience:** Content creators and faceless channel operators who want to post daily without the grind - **Pain Point Solved:** YouTube Shorts is printing money for channels that post daily. You cannot post daily because each video takes 4 hours. - **What Ships in the Box:** Complete Python automation pipeline, Gemini + ElevenLabs integration, auto-upload system, and cost breakdown docs. #### Key Features - Full pipeline: script → visuals → voiceover → captions → upload - Gemini AI for script writing, ElevenLabs for professional voice - Auto-captions that sync with the audio - Direct YouTube upload — zero manual steps - Cost breakdown: $0.11 per video in API fees #### Frequently Asked Questions - **Q: Is $0.11 per video really the real cost?** **A:** It is, for the default config: Gemini Flash for script, ElevenLabs Flash voice, stock visuals via Pexels, 30-45 second runtime. Switching to ElevenLabs premium voices pushes per-video cost to about $0.40. The cost breakdown doc shows every line item. - **Q: Will my YouTube channel get banned for automated uploads?** **A:** Not for using the YouTube Data API correctly with one channel and reasonable cadence (2-6 videos/day max). Mass-posting from multiple accounts is what triggers bans. The pipeline is built for one disciplined channel, not a spam farm. - **Q: What if I do not want to use my voice for visuals?** **A:** You do not — the default uses Pexels and Pixabay stock with motion overlays. Optional D-ID avatar integration is included if you want a synthetic talking head. Custom face uploads are not in v1. - **Q: How much daily volume can it actually handle?** **A:** On a standard machine with stable internet, the pipeline produces around 8-12 Shorts per hour end-to-end including upload. Daily volume is bottlenecked by your API quotas, not the script. Most operators settle at 4-6 published per day. - **Q: What niches work best?** **A:** Educational explainers, list videos, "did you know" facts, and curated news work great. Original storytelling and personality-driven channels do not — those need a human voice and face. The doc has a niche fit checklist. --- ### Startup Launch Arsenal ($39 USD) - **URL:** https://www.agentskillpacks.com/products/startup-launch-arsenal - **Category:** Business & Growth - **Type:** toolkit - **Tagline:** Go from idea to funded with 10 founder skills covering pitch, pricing, and launch. - **Summary:** A 10-skill founder toolkit that covers pre-launch and fundraising: investor pitch decks, pricing strategy using Van Westendorp, and the ORB launch framework. - **Target Audience:** Early-stage founders who need to pitch investors, price their product, and execute a launch - **Pain Point Solved:** You have a solid idea but your pitch is a mess, your pricing is a guess, and you have no idea how to launch. - **What Ships in the Box:** 10 founder skills, a pitch deck framework, the Van Westendorp pricing tool, and the ORB launch planning system. #### Key Features - 10 founder skills covering pitch, pricing, and launch strategy - Investor pitch deck structure with VC-approved slide order - Van Westendorp pricing: find the exact price your customers will pay - ORB launch framework: owned, rented, borrowed channels - From napkin idea to investor-ready in one structured system #### Frequently Asked Questions - **Q: Is this for fundraising or bootstrapping?** **A:** Both, but the pitch deck and investor briefing pieces lean toward fundraising. Bootstrapped founders use the pricing, launch, and ORB framework; the pitch material doubles as customer-facing deck once you remove the investor slides. - **Q: How specific is the pitch deck framework?** **A:** It mirrors the slide order that pre-seed and seed VCs expect: problem, insight, solution, why-now, market, business model, traction, team, ask. Each slide has a fill-in-the-blank template plus 2-3 examples from funded decks for context. - **Q: Does the Van Westendorp pricing tool need real customer data?** **A:** Yes — you need 20-30 survey responses minimum to get a usable price band. The toolkit ships with the survey template, the Google Form setup, and the analysis sheet. Without responses, the tool teaches the framework but cannot tell you a price. - **Q: What does the ORB launch framework cover?** **A:** Owned channels (newsletter, blog, podcast), Rented channels (social, ads, partners), Borrowed channels (PR, communities, influencers). The playbook prioritizes which to invest in first based on your team size and runway, with checklists for each. - **Q: Are the example decks redacted?** **A:** Most are anonymized — company names changed, numbers proxied. Two are public examples (early Stripe, early Notion-style structure) referenced for slide flow. Buyers should not expect raw confidential deck dumps. --- ### Structured Data Pro Pack ($19 USD) - **URL:** https://www.agentskillpacks.com/products/structured-data-pro-pack - **Category:** SEO & Visibility - **Type:** template - **Tagline:** Add the right schema markup. Show up in rich results and AI answers. - **Summary:** A JSON-LD schema library and implementation system covering Article, Product, Organization, FAQ, and speakable schemas that trigger Google rich results and AI citations. - **Target Audience:** SEO practitioners, developers, and site owners who want rich results without guessing at schema syntax - **Pain Point Solved:** Your content is good but it never shows up as a featured snippet or AI citation because the schema is missing or wrong. - **What Ships in the Box:** Full JSON-LD template library, schema implementation skill, validation workflow, and AI citation optimization guide. #### Key Features - Ready-to-use JSON-LD templates for Article, Product, and Organization - FAQPage schema that fires the Google rich result snippet - Speakable schema for AI assistant and voice search citations - Schema validation workflow — catch errors before Google does - Works with any CMS, framework, or static site #### Frequently Asked Questions - **Q: Will Google show rich results immediately after I add the schema?** **A:** No — Google has to recrawl, parse, and decide eligibility. Typical window is 3-21 days. FAQPage and Product are the fastest to appear. Article rich results appear less often even when valid because Google reserves the slot. - **Q: Does this work with WordPress, Webflow, Next.js, plain HTML?** **A:** All of them. The JSON-LD templates are framework-agnostic — paste into a script tag in any platform. The pack includes injection patterns for WP (no plugin), Webflow embed, and Next.js metadata API specifically. - **Q: Which schemas are in the library?** **A:** Article, Product, Organization, WebSite, Person, BreadcrumbList, FAQPage, HowTo, Review, AggregateRating, VideoObject, ImageObject, Speakable, Event, Recipe, LocalBusiness — 16 in total, with the most-cited variants per type. - **Q: Will valid schema improve AI citations?** **A:** Yes for ChatGPT, Perplexity, and Google AI Overviews — they all read JSON-LD for context. The bigger lever is page structure, but missing schema is the cheap fix that pays for itself in a week. The included guide ranks which schemas matter most for AI. - **Q: Is schema enough on its own?** **A:** No — schema describes what the page already says. If the content is thin or unclear, schema does not invent value. Pair this with the AI Content Blueprint for the full structure-plus-schema story. --- ### Visa Document Translator ($9 USD) - **URL:** https://www.agentskillpacks.com/products/visa-document-translator - **Category:** AI Tools - **Type:** toolkit - **Badge:** Popular - **Tagline:** Get your documents translated into 50 languages — in minutes, not weeks. - **Summary:** An OCR-powered document translator that reads visa documents, passports, birth certificates, and official papers and produces bilingual side-by-side translations with confidence flags. - **Target Audience:** Immigrants, expats, travelers, and international workers who need documents translated fast - **Pain Point Solved:** A certified translator costs $300 and takes 2 weeks. Your visa appointment is in 4 days and you have 6 documents. - **What Ships in the Box:** OCR translation tool, 50-language support, bilingual output format, confidence flag system, and cost comparison guide. #### Key Features - OCR document scanning: upload a photo and get text extracted - Bilingual side-by-side output format for easy review - Confidence flags that mark uncertain translations for human review - 50 languages supported for visa documents, birth certificates, passports - A fraction of the cost of a certified human translator #### Frequently Asked Questions - **Q: Are these translations accepted by embassies?** **A:** For initial application review, yes — embassies accept reference translations to read your documents. For sworn / notarized submission, no — you still need a certified human translator for the final stamp. This tool gets you 90% there for 3% of the cost. - **Q: Which 50 languages does it cover?** **A:** The 20 most common pairs (EN-ES, EN-FR, EN-DE, TR-DE, EN-AR, etc.) are tested production-quality. Another 30 are supported but with lower confidence scoring for legal-document jargon. Full list in the docs; check before buying if your pair is rare. - **Q: What document types does it handle?** **A:** Passports, birth certificates, marriage certificates, diplomas, transcripts, bank statements, tax documents, employment letters, police clearance certificates, visa application supporting documents. Handwritten documents are not supported — OCR fails on cursive. - **Q: How accurate is the OCR on scanned documents?** **A:** Clean scans (300 DPI, flat, well-lit) hit 97-99% character accuracy. Phone photos in good light are around 92-96%. Wrinkled, low-light, or angled photos drop to 80-90% and confidence flags catch most of the loss. The tool tells you when to rescan. - **Q: Is data stored anywhere after translation?** **A:** No — the tool runs on your machine via Ollama for the model and Tesseract for OCR. Document content never leaves your computer unless you opt into the optional GPT-4 high-accuracy mode. Default is fully offline. --- ## 2. Books & Series ### Solo Operator Series (Direct PDF + EPUB & Kindle) #### The Leverage Stack — How One Person Runs What Used to Take a Team - **URL:** https://www.agentskillpacks.com/books/the-leverage-stack - **Direct Edition:** $7 USD (PDF + EPUB + Workbook) - **Amazon Kindle URL:** https://a.co/d/02IgWyKZ - **Overview:** The Leverage Stack is a short, practical book for the solo operator who has become the bottleneck in their own business. It builds eight layers that catch the work that does not need your judgment, so only the decisions that genuinely need you reach you. Every chapter ends with a one-page tool you can use that day. - **Pitch:** You are not disorganized. You are the bottleneck — and you cannot out-discipline a structural problem. - **What You Learn:** Find the exact tasks that route only through you, and what to do with each one · Turn one-off AI use into a standing pipeline that runs without you · Trust delegated output without re-checking every line · Protect the work that must stay yours, so you do not automate away the only thing clients pay for · Ten one-page tools — an audit, a triage sheet, a reusable brief, a quality gate, a 30-day rollout ##### Chapter Breakdown & Tools - **Chapter 0 — Introduction — A Team Was Never the Goal:** Includes one-page tool *Leverage Audit* - **Chapter 1 — The Bottleneck:** Includes one-page tool *Single-Point Map* - **Chapter 2 — The Inventory:** Includes one-page tool *Task Triage Sheet* - **Chapter 3 — The Brief:** Includes one-page tool *Reusable Brief Template* - **Chapter 4 — The Pipeline:** Includes one-page tool *5-Step Pipeline Builder* - **Chapter 5 — The Quality Gate:** Includes one-page tool *Review Checklist + red-flag list* - **Chapter 6 — The Handoffs:** Includes one-page tool *Handoff Map* - **Chapter 7 — The Dashboard:** Includes one-page tool *One-Screen Ops Dashboard* - **Chapter 8 — The Human Layer:** Includes one-page tool *Keep / Delegate Line* - **Chapter 9 — Conclusion — Don’t Build the Whole Stack at Once:** Includes one-page tool *30-Day Stack Rollout* #### Found by AI — How Small Businesses Get Recommended by ChatGPT, Claude, Perplexity, and Google AI - **URL:** https://www.agentskillpacks.com/books/found-by-ai - **Direct Edition:** $7 USD (PDF + EPUB + Workbook) - **Amazon Kindle URL:** https://a.co/d/0aNkvqMm - **Overview:** Found by AI is a short, practical book for the small business owner who wants to be the name an AI recommends. It walks through the five questions ChatGPT, Claude, Perplexity, and Google AI quietly ask before they name you — can I find you, tell what you are, trust you, quote you, and reach you when asked. Each chapter ends with a one-page tool. - **Pitch:** You are not losing to a bigger ad budget. You are losing to a machine that does not know you exist. - **What You Learn:** Check whether ChatGPT, Claude, and Perplexity can read your site at all · Make it obvious to an AI what you do and who you serve · Build the trust signals an AI needs before it will name you · Write content an AI can quote cleanly, in passages that stand on their own · Seven one-page tools — a visibility snapshot, a crawler-access check, an entity sheet, a citability checklist, and a 30-day build ##### Chapter Breakdown & Tools - **Chapter 0 — Introduction — The New Front Door:** Includes one-page tool *AI Visibility Snapshot* - **Chapter 1 — Can I Even Find You?:** Includes one-page tool *AI Crawler Access Check* - **Chapter 2 — Can I Tell What You Are?:** Includes one-page tool *Entity Clarity Sheet* - **Chapter 3 — Do I Trust You Enough to Name You?:** Includes one-page tool *Trust Signal Map* - **Chapter 4 — Can I Quote You Cleanly?:** Includes one-page tool *Citability Checklist* - **Chapter 5 — Are You There When the Question Is Asked?:** Includes one-page tool *Prompt-Match Map* - **Chapter 6 — Conclusion — The 30-Day Visibility Build:** Includes one-page tool *30-Day Visibility Build* #### The One-Person Million — Build a Real Business With AI Doing the Work - **URL:** https://www.agentskillpacks.com/books/the-one-person-million - **Direct Edition:** $7 USD (PDF + EPUB + Workbook) - **Amazon Kindle URL:** https://a.co/d/0ewBC87Q - **Overview:** The One-Person Million is a short, practical book for the solo operator who is done trading time for money. It builds the structure behind large solo revenue one layer at a time — the offer, the price, the product, the AI-run delivery engine, and the stack of offers that sell each other. It is honest: most never reach a million, and it promises nothing. - **Pitch:** There are only so many hours in a week. If your income is chained to your presence, your ceiling was fixed the day you started. - **What You Learn:** Sell the outcome instead of the hour, so income stops tracking your time · Price for the value you create, not the minutes you spend · Turn a service into a product that earns while you sleep · Let AI run delivery so revenue is no longer capped by you · Seven one-page tools — a revenue-ceiling calculator, a value-price ladder, a delivery-engine map, and a 90-day build ##### Chapter Breakdown & Tools - **Chapter 0 — Introduction — The Ceiling of Hours:** Includes one-page tool *Revenue Ceiling Calculator* - **Chapter 1 — The Offer — Sell the Outcome, Not the Hour:** Includes one-page tool *Offer Reframe* - **Chapter 2 — The Price — Charge for the Value, Not the Time:** Includes one-page tool *Value Price Ladder* - **Chapter 3 — The Product — Make Money While You Sleep:** Includes one-page tool *Product Ladder* - **Chapter 4 — The Engine — Let AI Do the Delivery:** Includes one-page tool *Delivery Engine Map* - **Chapter 5 — The Stack — Offers That Sell Each Other:** Includes one-page tool *Offer Stack Map* - **Chapter 6 — Conclusion — The Million Is a Structure, Not a Sprint:** Includes one-page tool *90-Day Revenue Build* #### The Unfair Solopreneur — Compete With Companies 100x Your Size Using AI - **URL:** https://www.agentskillpacks.com/books/the-unfair-solopreneur - **Direct Edition:** $7 USD (PDF + EPUB + Workbook) - **Amazon Kindle URL:** https://a.co/d/0bqkXBfD - **Overview:** The Unfair Solopreneur is a short, practical book about the advantages that get sharper, not weaker, when you stay small and add AI. It walks through five edges a big company structurally cannot copy — speed, trust, focus, intimacy, and the firepower of running a whole department’s output solo. Each chapter ends with a one-page tool, and it promises no win, only the edges. - **Pitch:** You are not the underdog. You just think you are. - **What You Learn:** Ship while a bigger competitor is still scheduling the meeting · Turn being a real person, not a logo, into a trust advantage · Own the slice of the market too small for a giant to bother with · Run the output of a whole department, solo, with AI · Seven one-page tools — an advantage audit, a speed map, a trust checklist, and a 90-day build ##### Chapter Breakdown & Tools - **Chapter 0 — Introduction — The 100x Myth:** Includes one-page tool *Unfair Advantage Audit* - **Chapter 1 — Speed — You Move While They Meet:** Includes one-page tool *Speed Map* - **Chapter 2 — Trust — You’re a Person, They’re a Logo:** Includes one-page tool *Trust Edge Checklist* - **Chapter 3 — Focus — Own the Slice Too Small for Them:** Includes one-page tool *Niche Wedge* - **Chapter 4 — Intimacy — Know the Customers They Can Only Segment:** Includes one-page tool *Intimacy Map* - **Chapter 5 — Firepower — One of You, the Output of a Whole Department:** Includes one-page tool *AI Department Map* - **Chapter 6 — Conclusion — Press the Advantages While the Window Is Open:** Includes one-page tool *90-Day Unfair Build* #### AI Won’t Run Your Business. You Will. — Judgment and Doubt in the AI Age - **URL:** https://www.agentskillpacks.com/books/ai-wont-run-your-business - **Direct Edition:** $7 USD (PDF + EPUB + Workbook) - **Amazon Kindle URL:** https://a.co/d/01VnMx9f - **Overview:** AI Won’t Run Your Business is a short, practical book about the last human jobs in the AI age — the decisions, doubt, and judgment you must not hand to a machine, no matter how good it gets. AI does the what and the how; only you hold the why. Each chapter ends with a one-page tool. - **Pitch:** AI will draft your emails and sort your tasks. It will not decide what your business is for. - **What You Learn:** Own the decisions AI hands you, instead of deferring to a confident machine · Read your own doubt as information, not as weakness to suppress · Catch the confident AI in the moment it is wrong — the jagged frontier · Hold the why of your business, the one thing AI cannot do for you · Seven one-page tools — from a decision-ownership line to a 30-day judgment practice ##### Chapter Breakdown & Tools - **Chapter 0 — Introduction — The Hand on the Wheel:** Includes one-page tool *The Hand-on-the-Wheel Check* - **Chapter 1 — The Decision — AI Offers; You Choose, and You Live With It:** Includes one-page tool *Decision Ownership Line* - **Chapter 2 — The Doubt — Information to Read, Not Weakness to Suppress:** Includes one-page tool *Doubt Decoder* - **Chapter 3 — The Jagged Edge — Knowing When the Confident Machine Is Wrong:** Includes one-page tool *Judgment Gate* - **Chapter 4 — The Why — AI Does the What and the How; Only You Hold the Why:** Includes one-page tool *The Why Anchor* - **Chapter 5 — The Person You Become — Don’t Disappear Into the Tools:** Includes one-page tool *Agency Audit* - **Chapter 6 — Conclusion — The Machine Does the Work. You Do the Living.:** Includes one-page tool *30-Day Judgment Practice* ### Additional Kindle Books by İsmail Günaydın - **[From Zero to Sustainable Startup](https://a.co/d/0ejzri2t):** Building a business that survives its first year. Systems over hustle: how to pick a niche, validate it cheaply, and build the boring infrastructure that keeps a small business alive. Written from 15 years of shipped and failed projects. - **[Businesses Are Positive-Sum Games](https://a.co/d/01TpQiVn):** Scaling through coopetition instead of war. Why treating competitors as a market signal beats treating them as enemies. Covers partnerships, ecosystems, and growth loops that compound without ad spend. - **[Startup Spells: Loopholes & Leverage](https://a.co/d/01yslc6y):** Growth tactics for founders without budgets. A collection of leverage points solo founders can actually use — distribution shortcuts, positioning angles, and the small bets that occasionally pay off big. - **[The Startup Exploit](https://a.co/d/08v5oO0O):** Traction and speed for early-stage products. How early-stage products find their first users before the money runs out. Focused on speed of iteration, not vanity growth. - **[The Million Dollar Launch Manual](https://a.co/d/06Ut7UEz):** Launch mechanics, step by step. The anatomy of a product launch: sequencing, audience warm-up, and the checklist work nobody finds glamorous. The title oversells it; the checklists do not. - **[SEO Content, Rebuilt](https://a.co/d/0d459Dzm):** Data-driven SEO content in the AI era. How to write content that ranks on Google and gets cited by AI search at the same time. Covers intent mapping, content architecture, and what changed after AI Overviews. - **[The Bottom-of-Funnel SEO + LLM Cost Hacker's Guide](https://a.co/d/0aDjd1UD):** Enterprise SEO meets AI cost optimization. Two technical playbooks in one: capturing bottom-of-funnel search demand, and keeping LLM API bills sane while doing it at scale. - **[Reverse-Engineered Newsletters: The $10K/Month Blueprint](https://a.co/d/02SMC5NQ):** Newsletter business models, taken apart. Deconstructs how paid newsletters actually make money — sponsorships, paid tiers, products — using real published numbers. The subtitle is the genre talking; the analysis is sober. - **[Wealth Rewired: Escape the System. Build Real Freedom](https://a.co/d/08zjZxvS):** Rethinking money as a freedom budget. Less about getting rich, more about needing less: designing income, expenses, and work around autonomy instead of status. - **[No Borders, No Boss: How to Design a Life of Freedom](https://a.co/d/0006aBga):** Remote work and location independence. The practical side of location independence — income portability, residency, and the trade-offs the Instagram version leaves out. - **[Break Then Build](https://a.co/d/03JjoaHy):** Rebuilding after a failed project or career reset. For the moment after something collapses: how to audit what failed, keep what worked, and start the next thing without repeating the last one. - **[Don't Click That!](https://a.co/d/0gser3n7):** Recognizing online scams before they cost you. A field guide to phishing, fake shops, and AI-assisted fraud, written for non-technical readers. The defensive counterpart to everything else on this shelf. --- ## 3. Free Ungated Tools & Calculators Free, ungated web applications for digital sellers and builders. No login, no email required. ### Gumroad Fee Calculator - **URL:** https://www.agentskillpacks.com/gumroad-fee-calculator - **Description:** Itemizes Gumroad transaction fees, card processing cuts, merchant-of-record costs, and net payouts. ### Stripe Fee Calculator - **URL:** https://www.agentskillpacks.com/stripe-fee-calculator - **Description:** Itemizes Stripe transaction fees, card processing cuts, merchant-of-record costs, and net payouts. ### PayPal Fee Calculator - **URL:** https://www.agentskillpacks.com/paypal-fee-calculator - **Description:** Itemizes PayPal transaction fees, card processing cuts, merchant-of-record costs, and net payouts. ### Lemon Squeezy Fee Calculator - **URL:** https://www.agentskillpacks.com/lemon-squeezy-fee-calculator - **Description:** Itemizes Lemon Squeezy transaction fees, card processing cuts, merchant-of-record costs, and net payouts. ### Payhip Fee Calculator - **URL:** https://www.agentskillpacks.com/payhip-fee-calculator - **Description:** Itemizes Payhip transaction fees, card processing cuts, merchant-of-record costs, and net payouts. ### Etsy (digital downloads) Fee Calculator - **URL:** https://www.agentskillpacks.com/etsy-fee-calculator - **Description:** Itemizes Etsy (digital downloads) transaction fees, card processing cuts, merchant-of-record costs, and net payouts. ### Ko-fi Fee Calculator - **URL:** https://www.agentskillpacks.com/ko-fi-fee-calculator - **Description:** Itemizes Ko-fi transaction fees, card processing cuts, merchant-of-record costs, and net payouts. ### Additional Free Tools - **Refund Policy Generator:** https://www.agentskillpacks.com/refund-policy-generator — Generates EU CRD Art 16(m) & Directive (EU) 2023/2673 compliant digital refund terms. - **AI Crawler Checker:** https://www.agentskillpacks.com/ai-crawler-checker — Audits robots.txt against 12 major AI crawlers. - **llms.txt Generator:** https://www.agentskillpacks.com/llms-txt-generator — Builds valid llmstxt.org markdown specifications. - **AI Writing Checker:** https://www.agentskillpacks.com/ai-writing-checker — Flags 5 measurable AI style tells (em-dash density, stock phrases, rule-of-three). - **Digital Product Pricing Calculator:** https://www.agentskillpacks.com/digital-product-pricing-calculator — Recommends pricing bands and simulates monthly revenue. - **Platform Fee Comparison:** https://www.agentskillpacks.com/platform-fee-comparison — Compares net payouts across 7 payment platforms and 13 fee variants. --- ## 4. Editorial Blog Posts ### Claude Code skills for marketing: the three that survive contact with real work - **URL:** https://www.agentskillpacks.com/blog/claude-code-skills-for-marketing - **Date Published:** 2026-09-21 - **Tags:** claude-code, business-growth, ai-workflow, content-craft - **Excerpt:** Marketing skill packs promise forty workflows. Three categories survive real use, one fails in a way that costs you customers, and the reason is structural. - **Quick Answer:** Claude Code skills hold up in marketing where the work is structural: editing drafts against a style rule, producing variants of something already validated, and checking assets before they ship. They fail at the strategic layer, because a skill will produce a confident audience definition or positioning statement from nothing and you cannot tell it is wrong until the campaign is over. #### FAQs - **Q: What marketing work suits a Claude Code skill?** **A:** Structural work with a rule attached. Editing a draft against a banned phrase list, checking a page has an answer block under every heading, producing ten variants of a headline that already converted, validating metadata and social preview tags before publish. Each has a defined correct state. - **Q: Why do AI-written marketing drafts feel interchangeable?** **A:** Because the model optimises for the statistically safe phrasing, which is by definition the phrasing everyone else also landed on. That is not a prompt problem you can fix by asking for creativity. It is why the useful skills run after the draft exists rather than instead of writing it. - **Q: Should I use a skill to write blog posts?** **A:** Use one to edit them. Google rewards helpful content however it was produced, and separately targets mass-produced pages made to game rankings, so the risk is not authorship, it is sameness. A skill that catches sameness is worth more than one that produces it faster. - **Q: What about audience research and positioning skills?** **A:** Treat those as the highest-risk category. A skill will produce a fluent audience definition with no data behind it, and it looks identical to one built from customer interviews. The cost surfaces after the campaign, which is the worst possible feedback delay. - **Q: Do marketing skill packs replace a marketer?** **A:** No, and the packs that imply it are describing the part of the job that was never the bottleneck. They replace the passes a solo operator skips when tired, which is a smaller claim and a genuinely useful one. ### Claude Code skills for SEO: what to install and what not to automate - **URL:** https://www.agentskillpacks.com/blog/claude-code-skills-for-seo - **Date Published:** 2026-09-21 - **Tags:** claude-code, seo, developer-tools, ai-workflow - **Excerpt:** Which SEO jobs suit a Claude Code skill and which get worse when you automate them, plus the overlap problem in a 26-skill SEO install and how to fix it. - **Quick Answer:** Claude Code skills work well for SEO jobs with a checkable output: schema generation and validation, crawler and robots audits, internal link checks, metadata and heading structure. They work badly for keyword selection and competitive positioning, because both depend on judgement about a market the model cannot observe. Install for the first group, keep the second by hand. #### FAQs - **Q: What SEO tasks are worth turning into a Claude Code skill?** **A:** The ones with a verifiable output. Generating and validating JSON-LD, auditing robots.txt against current crawler documentation, checking that internal links resolve, reviewing heading hierarchy and metadata length. Each of those has a right answer that a machine can confirm, which is what makes automation safe. - **Q: What SEO work should stay manual?** **A:** Keyword selection, positioning against competitors, and deciding which page should own which query. All three depend on demand data and market judgement that the model does not have, and a skill will happily produce a confident answer anyway. A confident wrong keyword map is more expensive than no keyword map. - **Q: Can a skill replace an SEO tool subscription?** **A:** Partly. Skills replace the workflow and the checking, not the data. If your subscription is mostly for search volume and backlink indexes, a skill does not substitute for it. If it is mostly for audits and checklists you could run yourself, it largely does. - **Q: Why do my SEO skills keep picking the wrong one?** **A:** Because SEO skills overlap more than almost any other domain. Technical audit, on-page review, schema and content scoring all sound similar in a one-line description. Rewrite the descriptions against each other so each one names its boundary. - **Q: Does any of this help with AI search specifically?** **A:** The crawler access and schema parts do, because both decide whether AI systems can read and parse you at all. The content side is a separate discipline, and it is editorial rather than technical. ### Free vs paid Claude Code skills: what the free ones actually cost - **URL:** https://www.agentskillpacks.com/blog/free-vs-paid-claude-code-skills - **Date Published:** 2026-09-21 - **Tags:** claude-code, developer-tools, ai-workflow, solo-founder - **Excerpt:** Most Claude Code skills are free and that is usually the right answer. Measured numbers from a 383-skill install on what the free route costs in context. - **Quick Answer:** Most Claude Code skills are free. You install them from public GitHub repos with no payment and no account, and for a lot of work that is the correct choice. The cost is not money: every installed skill puts its description in front of the model when it decides what to do, so a large free install makes that decision worse, and unmaintained skills fail silently. #### FAQs - **Q: Are Claude Code skills free?** **A:** Most of them are. A skill is a folder with a SKILL.md file in it, and public repositories full of them exist on GitHub at no cost. Paid packs also exist. Both install the same way and neither has a technical advantage over the other, so the question is curation and maintenance rather than capability. - **Q: Do more installed skills make Claude Code better?** **A:** No, and past a point the relationship reverses. Every installed skill contributes its description to the decision about which skill to use, so a large install spends context on routing and gives the model more near-identical options to choose between. Curation is the feature. - **Q: What should I pay for, if anything?** **A:** Curation, maintenance and enforcement. A pack is worth money when it removes a choice you would otherwise make badly, keeps working after the CLI changes, and refuses to continue when you skip a step. Capability alone is not worth paying for, because the free repos have it. - **Q: How do I tell a good skill from a bad one before installing?** **A:** Read the description field. If it does not say when the skill should fire, the model has nothing to match your request against and the skill will sit unused. Then check the repository's last commit date against the last Claude Code release that changed skill behaviour. - **Q: Is there a downside to installing a large free pack just in case?** **A:** Yes, and it is the one nobody mentions. Skills you never use still cost you on every request that has to choose between them, and they crowd out the ones you do use. An unused skill is not neutral; it is noise with a folder around it. ### Where to find Claude Code skills in 2026: repos, directories and packs - **URL:** https://www.agentskillpacks.com/blog/where-to-find-claude-code-skills - **Date Published:** 2026-09-21 - **Tags:** claude-code, developer-tools, ai-workflow - **Excerpt:** The four places Claude Code skills come from, what each one is good for, and the maintenance question that decides whether a skill still works next month. - **Quick Answer:** Claude Code skills come from four places: Anthropic's own repository, public GitHub repos from individual developers, aggregator directories that index those repos, and paid packs. All four install identically by dropping a folder into your skills directory. The difference is curation and whether anyone maintains them after the CLI changes. #### FAQs - **Q: Where do Claude Code skills come from?** **A:** Four sources. Anthropic's own public repository, individual developers publishing repos on GitHub, aggregator directories that index those repos, and paid packs sold directly. All of them produce the same thing, a folder containing a SKILL.md file, and all install the same way. - **Q: Is there an official Claude Code skills marketplace?** **A:** There is an official repository from Anthropic, which is the reference for the format. The sites describing themselves as marketplaces are mostly indexes over public GitHub repositories, which is useful for discovery and is not the same as vetting. - **Q: How do I install a skill I found?** **A:** Copy the skill's folder into your skills directory, either the user-level one for everything you do or the project's own for a single repository. The CLI picks it up from there. Restart if it does not appear. - **Q: How do I tell whether a skill is maintained?** **A:** Check the last commit date against the most recent Claude Code release that changed anything about skills. A repo that predates a format change will usually fail silently rather than error, which is the worst kind of failure to debug. - **Q: How many should I install at once?** **A:** Three, then use them for a week. Skills compete for selection, so a bulk install degrades the behaviour of everything already in the folder. The instinct to grab fifty at once is the single most common mistake here. ### Why half your Claude Code skills never fire (and how to fix the descriptions) - **URL:** https://www.agentskillpacks.com/blog/why-claude-code-skills-dont-trigger - **Date Published:** 2026-09-21 - **Tags:** claude-code, developer-tools, ai-workflow - **Excerpt:** Claude Code picks a skill by reading descriptions. In a 383-skill install only 136 said when to fire. The rule that decides it, and how to rewrite the rest. - **Quick Answer:** Claude Code selects a skill by matching your request against each skill's description, so a description that says what the skill is rather than when to use it gives the model nothing to match. In a 383-skill install I measured, only 136 descriptions named a trigger condition. The fix is rewriting the description, not the skill. #### FAQs - **Q: Does Claude Code use skills automatically?** **A:** It can, and whether it does depends almost entirely on the description field. Claude Code reads the available skill descriptions and matches your request against them, so a description that explains what a skill is without saying when it applies will usually lose to one that names the situation. You can also invoke a skill by name. - **Q: Why does my skill never get picked?** **A:** Most often because its description describes the skill rather than the trigger. "A comprehensive testing workflow" tells the model what it is. "Use when the user asks to add tests, fix a failing test, or raise coverage" tells the model when to reach for it. The second one fires. - **Q: Do overlapping skills interfere with each other?** **A:** Yes. If several skills plausibly match a request, the model has to pick one, and near-identical descriptions make that a coin flip. Skills in the same domain need descriptions written against each other, each naming what it handles and what it leaves to its neighbour. - **Q: Can subagents use skills?** **A:** A subagent works from the tools and instructions it is given, so treat skill availability as something to state explicitly in the delegation rather than something to assume. If a subagent must follow a workflow, name the skill in its prompt instead of hoping selection goes your way. - **Q: How many skills is too many?** **A:** There is no hard limit, and the practical ceiling arrives sooner than people expect because selection quality degrades before anything breaks. If you cannot name what each installed skill does and when it fires, you are past it. ### Does llms.txt actually work? What the server logs say in 2026 - **URL:** https://www.agentskillpacks.com/blog/does-llms-txt-actually-work - **Date Published:** 2026-09-19 - **Tags:** geo-seo, ai-search, llms-txt - **Excerpt:** Google says no AI system uses llms.txt. A 137,210-domain log study found 97% of the files got zero requests. Who does read them, and whether to ship one. - **Quick Answer:** Not for AI search visibility, going by the public evidence in 2026. Google's AI features guide says you need no AI text files to appear in AI Overviews, and an Ahrefs log study of 137,210 domains found 97% of llms.txt files got zero requests in May 2026. Coding agents like Claude Code do read the file, and it takes minutes to write, so ship it as a courtesy, not as a tactic. #### FAQs - **Q: Does Google use llms.txt for AI Overviews?** **A:** No. Google's AI features guide says you do not need to create new machine-readable files, AI text files, or markup to appear in its AI features. Being crawled and indexed by Googlebot like any other page is the requirement. - **Q: Which AI tools actually read llms.txt?** **A:** Mostly not AI tools at all: in Ahrefs' May 2026 data, 77% of the bots fetching llms.txt were SEO auditors, profilers, and other crawlers. Among AI tools, coding agents stand out. Claude Code out-fetched every AI retrieval bot, assistant, and training crawler except GPTBot and one agent-infrastructure bot. - **Q: Is llms.txt an official standard?** **A:** No. It is a proposal Jeremy Howard published at llmstxt.org on September 3, 2024. No standards body has adopted it, and no major search engine has committed to reading it. - **Q: Should I delete my llms.txt?** **A:** No need. It costs nothing to serve and helps developers who point a coding agent at your site. Just stop counting it as an AI visibility tactic, and put the hours into crawlable pages and quotable passages. ### Gumroad vs Payhip vs Etsy vs Ko-fi vs Lemon Squeezy: which is better for digital downloads? - **URL:** https://www.agentskillpacks.com/blog/gumroad-vs-payhip-vs-etsy-vs-ko-fi-vs-lemon-squeezy - **Date Published:** 2026-09-19 - **Tags:** pricing-strategy, payment-infrastructure, solo-founder - **Excerpt:** Net payout on a $9, $29 and $119 digital download across five platforms, what each fee includes, who handles tax, and which one fits which seller. - **Quick Answer:** On a $29 digital download sold to a US buyer, Lemon Squeezy leaves you $27.05, Payhip's free plan and Ko-fi's free tier $26.41, Etsy $25.79, and Gumroad $24.46 on a direct sale. Gumroad Discover and Etsy Offsite Ads sales drop to about $20-21. Lemon Squeezy and Gumroad also act as merchant of record, so they handle sales tax and VAT for you. #### FAQs - **Q: Which is better, Gumroad or Payhip?** **A:** Payhip, on fees alone. Its free plan charges 5% plus card processing, so a $29 sale nets $26.41 against $24.46 on Gumroad direct. Gumroad's edge is Discover, its marketplace, and the fact that it handles sales tax and VAT as merchant of record. - **Q: Which is better, Gumroad or Etsy, for digital products?** **A:** Etsy keeps less of each sale than it looks, but it brings buyers who are already searching. A standard $29 Etsy sale nets $25.79, more than Gumroad direct. A sale Etsy attributes to Offsite Ads drops to $21.44. Pick Etsy for products shoppers browse for, such as printables and planners. - **Q: Is Lemon Squeezy owned by Stripe?** **A:** Yes. Stripe acquired Lemon Squeezy on July 26, 2024. It still runs as a merchant of record at 5% + $0.50 per transaction, with card processing included in that fee. - **Q: Which is better, Gumroad or Lemon Squeezy?** **A:** Lemon Squeezy on fees: its 5% + $0.50 includes card processing, so a $29 sale nets $27.05 against $24.46 on Gumroad direct. Both act as merchant of record and handle sales tax. Gumroad's edge is Discover, a marketplace that can send buyers you did not find yourself. - **Q: When is a paid plan worth it?** **A:** When your monthly unit volume clears the plan's crossover. At a $29 price, Payhip Plus at $29 a month beats the free plan from 34 sales a month, Payhip Pro at $99 needs 69, and Ko-fi Gold at $12 needs 9. At a $9 price every one of those thresholds roughly triples. - **Q: Which platform is free to sell digital products on?** **A:** Gumroad, Payhip's Free Forever plan, Ko-fi's free tier, and Lemon Squeezy have no monthly fee; they take a cut per sale instead. Etsy charges $0.20 per listing. Payhip's paid plans and Ko-fi Gold swap part or all of the per-sale fee for a monthly charge. ### Is your site visible to AI crawlers? The 5-minute audit - **URL:** https://www.agentskillpacks.com/blog/ai-crawler-access-5-minute-audit - **Date Published:** 2026-06-10 - **Tags:** geo, seo, ai-search - **Excerpt:** A copy-pasted 2023 robots.txt block can erase you from ChatGPT and Perplexity. The 5-minute audit, the bots that matter, and the allow-search strategy. - **Quick Answer:** Audit AI crawler access in three steps. Fetch yourdomain.com/robots.txt and check rules against the 12 AI bots that matter (GPTBot, OAI-SearchBot, ClaudeBot, PerplexityBot, Google-Extended, CCBot, and friends). Decide separately for search bots versus training bots. Then give allowed crawlers a curated map via llms.txt. No robots.txt at all means everything is allowed, which for most sellers is the correct state. #### FAQs - **Q: Does blocking GPTBot remove my site from ChatGPT answers?** **A:** Not directly. GPTBot collects training data; ChatGPT's live answers with citations come through OAI-SearchBot and ChatGPT-User. Blocking GPTBot keeps you out of future training corpora while leaving you citable. Blocking the search bots is what removes you from answers, and traffic. - **Q: Does blocking Google-Extended hurt Google rankings?** **A:** No. Google has documented that Google-Extended only controls Gemini model training. It is not a Search ranking signal, and AI Overviews use Googlebot. You can block Google-Extended and lose nothing in classic search results. - **Q: My site has no robots.txt at all. Is that a problem?** **A:** No file means every compliant crawler assumes full access, which for most product and content sites is exactly the right state. It is the sites that copy-pasted a block-everything snippet during the 2023 panic that need the audit most, because they opted out of AI answers without deciding to. - **Q: Do AI companies actually respect robots.txt?** **A:** The documented major bots do: OpenAI, Anthropic, Google, and Apple publish their user agents and honor the protocol. Reported violations cluster around smaller undisclosed scrapers. robots.txt is a published preference, not a wall; for hard enforcement you need CDN or WAF rules. ### Gumroad fees in 2026: what sellers actually pay per sale - **URL:** https://www.agentskillpacks.com/blog/gumroad-fees-2026-actually-itemized - **Date Published:** 2026-06-10 - **Tags:** pricing-strategy, solo-founder, business-growth - **Excerpt:** Gumroad says 10%. Your payout says otherwise. Every fee on a direct and a Discover sale, the price floor where you lose money, and when leaving pays. - **Quick Answer:** Gumroad takes 10% + $0.50 per direct sale, and card processing of 2.9% + $0.30 is billed on top, so the real rate is about 12.9% + $0.80. A $29 product nets $24.46. Sales through Gumroad Discover pay a flat 30% instead, netting $20.30. Free products cost nothing, and Gumroad has handled sales tax as merchant of record since January 1, 2025. #### FAQs - **Q: How much does Gumroad take per sale?** **A:** On a direct sale Gumroad takes 10% + $0.50, and card processing of 2.9% + $0.30 is charged separately. On a $29 product the combined deduction is $4.54, an effective 15.7%, and you keep $24.46. The 10% headline is true and incomplete at the same time. - **Q: Does Gumroad have a monthly plan that lowers the fee?** **A:** Not since 2023. Gumroad ran tiered subscriptions years ago, the last of them a $10 per month plan, and retired all of them in favour of a single flat rate. Its pricing page now states plainly that Gumroad does not charge a monthly fee. Articles quoting a $10 plan at 3.5% are describing a product that no longer exists. - **Q: What does Gumroad Discover cost?** **A:** A flat 30% of the sale, processing included, whenever the buyer arrives through Gumroad's marketplace search or recommendations instead of your own link. There is no fixed fee on Discover sales, which makes it the cheaper route for products priced under about $4.68. - **Q: Does Gumroad charge for free products?** **A:** No. The 10% + $0.50 fee only applies when money changes hands, so a $0 lead magnet or free download costs the seller nothing. That makes Gumroad a reasonable host for a free sample even if you sell the paid version somewhere else. - **Q: Does Gumroad handle sales tax and VAT?** **A:** Yes. Gumroad became the merchant of record for its sellers on January 1, 2025, which means it collects and remits sales tax, VAT, and GST on your behalf. That is included in the fee, and it is the main thing you give up by running your own checkout. - **Q: At what price does a Gumroad sale lose money?** **A:** Below roughly $0.92 on direct sales. The fixed fees ($0.50 platform plus $0.30 processing) and the percentages exceed the sale price at that point. A $0.99 product nets about six cents. - **Q: When is it worth leaving Gumroad for your own checkout?** **A:** The gap versus self-hosted Stripe is 10 percentage points plus $0.50 per sale. At $500 a month that is roughly $60, not worth the build. Around $3,000 a month it reaches about $350, which can fund your own checkout, as long as you also take on VAT registration and Stripe Tax. ### How to price a digital product (with the calculator math shown) - **URL:** https://www.agentskillpacks.com/blog/how-to-price-a-digital-product-calculator-math - **Date Published:** 2026-06-10 - **Tags:** pricing-strategy, solo-founder, business-growth - **Excerpt:** The three-dial pricing model behind our free calculator: market bands, audience and depth multipliers, ladder snapping, and the fee step most models skip. - **Quick Answer:** Price a digital product from three inputs. Start with the market band for the product type (templates typically $19-$79, courses $29-$299), multiply by an audience factor (hobbyists 0.7x, professionals 1.0x, businesses 1.6x) and a depth factor (starter 0.8x, comprehensive 1.4x), then snap to a 9-ending price. Model revenue at 0.5%, 1%, and 2% conversion, after fees. #### FAQs - **Q: Why do businesses pay 1.6x what hobbyists pay for the same product?** **A:** Different comparison anchors. A business buying a $79 template compares it against an employee-hour, roughly $50-150 of fully loaded cost. A hobbyist compares it against a streaming subscription. The same artifact sits in two different mental budgets, and the multiplier shows up consistently wherever both segments buy the same kind of product. - **Q: Should digital product prices end in 9?** **A:** Catalogs overwhelmingly do, and field experiments going back to Anderson and Simester's 2003 catalog tests found 9-endings lifted demand, in one case selling more units at $39 than at $34. The more useful detail is ladder spacing. Gaps should widen as prices rise, because buyers perceive price differences as ratios rather than dollars. - **Q: What conversion rate is realistic for a digital product?** **A:** Around 0.5% from cold traffic, 1% from an engaged audience, and 2% or better from a warm email list. Published ecommerce conversion averaged 2.23% in August 2026, and that figure is lifted by returning buyers on established stores. Run the revenue math at 0.5% and make sure the pessimistic case still clears your effort bar. - **Q: Is it ever right to price below $9?** **A:** Only for genuine impulse printables sold in volume. Below $9, fixed payment fees eat 10-35% of the sticker on most platforms, and price itself signals quality where buyers cannot inspect the product first. Bundling three $4 items into one $12 product fixes both problems at once. - **Q: Should I price from my costs?** **A:** No. A digital product has almost no marginal cost, so cost-plus pricing on a product that took three weeks to build produces a number that says more about your typing speed than about the buyer. Price from the band the market already accepts and the budget the buyer draws from. ### The 7 AI writing tells editors catch first - **URL:** https://www.agentskillpacks.com/blog/seven-ai-writing-tells-editors-catch - **Date Published:** 2026-06-10 - **Tags:** content-and-writing, ai-workflow, solo-founder - **Excerpt:** Em-dash overload, "delve into", autopilot triads, same-length sentences. The tells editors catch first, why detectors miss the point, and our own checklist. - **Quick Answer:** The seven tells editors catch first: em-dash density above 4 per 1,000 words, stock phrases like "delve into" and "moreover", rule-of-three padding, uniform sentence length (variation under 35%), zero first-person voice in long copy, hedging without stakes, and conclusions that restate the introduction. Five of the seven are mechanically measurable. #### FAQs - **Q: Can these tells prove a text was written by AI?** **A:** No, and anyone selling that certainty is overselling. Humans trained on corporate and academic writing produce the same flat patterns, and a careful AI-assisted writer produces none of them. The tells predict whether a reader's trust survives, which matters commercially whether or not a model touched the draft. - **Q: Why are AI detectors unreliable?** **A:** The same paragraph routinely scores "98% AI" on one detector and "human" on another, false accusations against students are documented, and every model release resets the arms race. Detection guesses at authorship. Editing measures the prose itself, which is both checkable and fixable. - **Q: What is burstiness in writing?** **A:** The variation between short and long sentences. Models regress toward medium-length sentences; people interrupt themselves, fragment, then run on. Measured as the coefficient of variation of sentence lengths, anything under about 35% reads mechanically flat. - **Q: Is using AI to draft content bad for SEO?** **A:** Google's published position targets scaled low-value content, not AI assistance itself. The practical risk is subtler: machine-flat prose loses readers by paragraph three, and engagement losses compound into ranking losses. Draft with AI if it helps. Edit until the tells are gone. ### Best Claude Code skills for indie developers (with real examples) - **URL:** https://www.agentskillpacks.com/blog/best-claude-code-skills-for-indie-developers - **Date Published:** 2026-06-03 - **Tags:** claude-code, developer-tools, solo-founder, ai-workflow - **Excerpt:** The Claude Code skills that move real work: TDD, planning, code review, multi-agent. Six skill packs solo developers use and the gap each one closes. - **Quick Answer:** The best Claude Code skills for indie developers are the ones that enforce a workflow rather than write code in a vacuum. CodeForge (14 skills covering TDD, planning, code review, git worktrees) is the foundational pack. Add one specialist pack per discipline: Indie Game Studio Kit for game dev, AI SEO Command Suite for SEO ops, AI Agency Launch Kit for client work. Six packs cover most solo-developer use cases; you build the rest by composing them. #### FAQs - **Q: What are Claude Code skills exactly?** **A:** Claude Code skills are reusable workflow definitions that drop into `~/.claude/skills/` and run inside the Claude Code CLI. Each skill describes a job (run TDD, plan a feature, review a PR) and the steps the agent should take. Skills compose, so one skill can call another and a single project can carry several. - **Q: Why use skills instead of just chatting with Claude?** **A:** Chatting works for one-off prompts. Skills enforce a repeatable workflow you cannot easily skip: TDD red-green-refactor, planning before code, automatic code review after edits. Solo developers benefit most because skills replace the discipline a senior teammate would normally enforce. The first week feels slower; from week two on, rework drops sharply. - **Q: Do skills require a paid Claude subscription?** **A:** Yes. Claude Code runs on the paid Claude API or a Pro or Max subscription. Costs vary by model and volume. The skills themselves are tool-agnostic logic; many have adapter configs that work in Cursor (custom instructions), Codex CLI, and Gemini CLI for 80-85% of the value at different cost structures. - **Q: Where do indie developers find Claude Code skills worth installing?** **A:** Three sources stand out. ToolGenX ships six paid skill packs from one maintainer (CodeForge, Indie Game Studio Kit, AI SEO Command Suite, AI Agency Launch Kit, Humanizer Pro, AI Content Empire Builder). The official Anthropic skills repo on GitHub has free starter skills. Individual solo developers publish skills on their personal sites and through Gumroad. - **Q: How many skills should I install before it gets messy?** **A:** Around 30-50 active skills is the practical ceiling on a single Claude Code instance before discoverability suffers. Most solo developers run 15-25 active skills covering their primary workflow plus 3-5 specialist agents. The Claude Code CLI now has skill filtering, which raises the ceiling, but quality beats quantity. - **Q: Can Claude Code skills work without internet?** **A:** The skills themselves are plain text and run locally. The Claude API calls they make require internet. If you need fully offline workflow agents, Ollama-based setups (covered in AutoEarnaAI) are the closest indie option, though the model quality is below Claude for code. ### Is Gumroad worth it in 2026? What a year on a redirect checkout teaches you - **URL:** https://www.agentskillpacks.com/blog/a-year-on-gumroad-19-products-one-sale - **Date Published:** 2026-05-28 - **Tags:** solo-founder, postmortem, payment-infrastructure - **Excerpt:** Gumroad is worth it at one stage and quietly expensive at the next. The stage table, the blind spot a redirect creates, and the test that tells you which. - **Quick Answer:** Gumroad is worth it for validating a first digital product: zero setup, free products cost nothing, and it handles sales tax as merchant of record. It stops being worth it when your question changes from "will anyone buy" to "why are they not buying", because a redirect checkout hides every step between the buy button and the receipt. #### FAQs - **Q: Is Gumroad worth it in 2026?** **A:** For a first product, yes. You can list in minutes, $0 products cost nothing, and since January 2025 Gumroad files your sales tax and VAT. It is less worth it once you have steady traffic, because the 12.9% + $0.80 real fee is high and the checkout gives you little funnel data. - **Q: What is the real cost of a redirect checkout?** **A:** Diagnosis time. When the buy button sends people to another domain you can see the click and the completed order, and nothing in between. If conversion is low you cannot tell whether the offer, the price, the checkout form, or the handoff itself is responsible, so every fix is a guess you have to wait a month to evaluate. - **Q: How do I tell whether my offer or my funnel is the problem?** **A:** Compare buy-button clicks on your own site against completed sales on the platform. A high click rate with almost no completions points at the funnel. A low click rate on healthy traffic points at the offer or the price. Until you have both numbers you are choosing between two explanations with no evidence. - **Q: Can I instrument the funnel without leaving Gumroad?** **A:** Yes, and it is the cheapest experiment available. Fire an analytics event on every buy-button click, then reconcile that count against Gumroad sales weekly. It will not show you what happens inside Gumroad's checkout, but it tells you whether the drop is before or after the handoff, which is the fork that decides everything else. - **Q: When should I move off Gumroad?** **A:** Set the trigger before you need it. A count works better than a feeling: move to your own checkout at a fixed number of cumulative sales, or once the monthly fee gap funds the build. The fee gap reaches roughly $350 a month at $3,000 in sales, which is when the arithmetic starts to agree with the instinct. - **Q: Would you recommend Gumroad for a first product?** **A:** Yes. Validate the offer with zero engineering and zero compliance work. The mistake is not starting there. The mistake is leaving it as the primary checkout long after the data has started asking a question the platform cannot answer. ### How I rewrote 19 product descriptions in 2 days with humanizer-pro - **URL:** https://www.agentskillpacks.com/blog/rewriting-19-product-descriptions-in-2-days-with-humanizer - **Date Published:** 2026-05-21 - **Tags:** content-craft, humanizer, ai-search, solo-founder - **Excerpt:** The workflow that takes AI-drafted product copy through a humanizer pass so it reads like one person wrote it and gives AI engines something to quote. - **Quick Answer:** Rewriting 19 generic AI-drafted product descriptions in two days takes a three-pass workflow: first pass removes 50 banned AI phrases (transform, unlock, in todays), second pass brings em-dash density under 4 per 1000 words and breaks the uniform rhythm, third pass injects first-person voice and one specific number per major section. Batching by pass rather than by page is what makes two days possible. #### FAQs - **Q: How long does each description actually take after the workflow is set up?** **A:** About six minutes per product page once the workflow is muscle memory. Twelve in the first hour, six by the third. Nineteen products is roughly two focused hours plus a coffee break. The setup pass to identify the banned phrases and voice samples is what takes longer the first time. - **Q: Will ChatGPT and similar tools detect AI-edited prose as AI?** **A:** Detection rates drop sharply once the structural tells are removed. The surface signals (banned phrases, em-dash bursts, rule-of-three list padding) are the easy stuff. After a humanizer pass those are gone. What remains is harder to flag because it just reads like writing. - **Q: Does this change the SEO of the rewritten pages?** **A:** Yes, in two specific ways. First, AI search engines (ChatGPT, Perplexity, Gemini) cite distinctive passages, and humanized prose has more of those per page. Second, Google's helpful content classifier weighs against template-feeling copy, so the rewrite reduces a soft penalty risk. - **Q: Is there a free version of this workflow?** **A:** The three-pass framework itself is free, and it is what is in this post. The skill that automates it (banned phrase list, voice matching, before and after comparison) is the humanizer-pro product. You can do it manually with a spreadsheet of phrases to find-replace if you have the patience. - **Q: What if I have hundreds of product pages, not 19?** **A:** Two strategies. Either prioritize the top 20% by traffic and rewrite those first (Pareto applies). Or batch-process via the humanizer skill in Claude Code which can chew through a directory of MDX files in a single run. I would still spot-check every output by hand though. ### Building the EU CRD Article 16(m) consent flow in Next.js - **URL:** https://www.agentskillpacks.com/blog/eu-crd-article-16m-consent-flow-in-nextjs - **Date Published:** 2026-05-14 - **Tags:** legal, payment-infrastructure, solo-founder, eu-compliance - **Excerpt:** The EU CRD Article 16(m) waiver for digital goods in Next.js. The dual-checkbox UI, the consent_logs schema, and the Withdrawal Button required since June 2026. - **Quick Answer:** EU CRD Article 16(m) lets a digital goods seller waive the buyer's 14-day right of withdrawal only if the buyer expressly requests immediate access AND acknowledges losing the right to withdraw. In a Next.js checkout this is two unticked checkboxes, both gating the buy button, with consent text, IP, user-agent, and timestamp written to a consent_logs row for 6 years. From June 19, 2026 a Withdrawal Button is also mandatory. #### FAQs - **Q: Does this apply if I am not in the EU but my buyers are?** **A:** Yes. The directive protects EU consumers regardless of the seller's location. If a buyer in Germany buys from a Turkish shop, German consumer law including the CRD applies to that contract. The same is true for any EU member state buyer. - **Q: What happens if I do not collect the dual consent?** **A:** The right of withdrawal stretches from 14 days to 14 days plus one year. If the buyer requests a refund inside that extended window, you owe it, even if the file was downloaded immediately. The burden of proof that consent was given is on the seller. - **Q: Is one checkbox covering both points enough?** **A:** No. The directive requires the two consents to be separate and given individually. One combined checkbox is interpreted as bundling, which makes the waiver invalid. Two checkboxes, both unticked by default, both required. - **Q: Does the Withdrawal Button apply to digital goods that were downloaded?** **A:** The button itself must be visible on every eligible order. Eligibility means inside the 14-day window AND the right of withdrawal still applies. For downloaded digital goods where consent was properly captured, the button shows a disabled state with the reason. For undownloaded orders the button is active. - **Q: What is the simplest schema for consent_logs?** **A:** A single table with columns id, user_id, email, consent_type (immediate_access or withdrawal_waiver), consent_text_hash, consent_text_version, ip, user_agent, locale, created_at. One row per consent moment. Retention 6 years matches both the EU statute of limitations and the typical legal claim window. ### Shipping a digital product as a solo founder - **URL:** https://www.agentskillpacks.com/blog/shipping-a-digital-product-as-a-solo-founder - **Date Published:** 2026-05-12 - **Tags:** solo-founder, digital-products, payment-infrastructure, geo-seo - **Excerpt:** A playbook from one founder running a 19-product shop. Distribution, payment infrastructure, refund law, content gravity, and trust signals that move money. - **Quick Answer:** Shipping a digital product as a solo founder means owning four things end-to-end: distribution (one channel, repeated), payment (a platform like Gumroad first, your own gateway once volume pays for the plumbing), legal (EU CRD-compliant refund flow with consent logs), and content gravity (a small site that gets cited by Google AND ChatGPT). The rest is finishing what you started. #### FAQs - **Q: How long should a solo founder spend on the site before working on distribution?** **A:** Two to four weeks of focused build, no more. After that, every week spent polishing the site is a week not spent learning what your audience actually wants. I rebuilt ToolGenX in three weeks and the very next day I went back to writing. - **Q: Should I sell through Gumroad or build my own checkout?** **A:** Sell through Gumroad for the first three to ten sales. The moment the math says the platform fee costs more than a weekend of checkout wiring, switch. I built my own checkout before reaching that point, then switched it off and went back to Gumroad. Your number will be different. - **Q: How big does the catalog need to be?** **A:** One product is enough to validate. Ten is enough to look like a real shop. Nineteen is what I happened to have when I rewrote. The math is not about catalog size, it is about whether any single product is actually wanted. - **Q: Do I need a blog?** **A:** Yes, but not the kind you are thinking of. You need three to ten posts that answer the exact questions your buyer types into Google or ChatGPT before they decide. Not a content factory. Not weekly schedules. Just real answers to real questions, kept alive. - **Q: What kills more solo digital product shops than anything else?** **A:** Building features instead of selling. The second-biggest killer is a silent failure mode such as a redirect checkout, where you cannot see who reached the payment step or where they stopped. Instrument that half of the funnel before you optimise anything downstream of it. ### ToolGenX's robots.txt and llms.txt explained, line by line - **URL:** https://www.agentskillpacks.com/blog/toolgenx-robots-and-llms-txt-explained - **Date Published:** 2026-05-07 - **Tags:** geo-seo, ai-search, solo-founder - **Excerpt:** The robots.txt and llms.txt files on ToolGenX, line by line. What each rule does, why AI crawlers are allowed, and how llms.txt actually gets used. - **Quick Answer:** The robots.txt at ToolGenX explicitly allows GPTBot, OAI-SearchBot, ChatGPT-User, ClaudeBot, PerplexityBot, Google-Extended, CCBot, Applebot-Extended, and Bytespider, which is all the major AI crawlers, because the site is built to be cited in AI search answers. The llms.txt at the root is a curated 200-line markdown tour of the catalog, prices, and policies that AI assistants can read in one fetch instead of crawling 38 HTML pages. #### FAQs - **Q: Is llms.txt actually used by AI assistants today?** **A:** Adoption is uneven but climbing. Anthropic and Perplexity have referenced the spec in changelogs. Google has not committed publicly. OpenAI has not committed publicly. Smaller AI search startups have adopted it eagerly. The file takes 10 minutes to write and breaks nothing, so the question is academic, so just ship it. - **Q: Will allowing GPTBot hurt my Google ranking?** **A:** No. GPTBot trains OpenAI models and powers ChatGPT browsing. Googlebot is a separate crawler that handles Google Search indexing. Blocking one does not affect the other. The two operate completely independently. - **Q: What is the difference between GPTBot and ChatGPT-User?** **A:** GPTBot crawls broadly to train the underlying model. ChatGPT-User fetches pages on demand when a user asks ChatGPT to browse or search. Blocking one without the other gives a half-result: you can be in the training corpus but not citable in real-time browsing, or vice versa. - **Q: Should I block Google-Extended?** **A:** Only if you actively do not want Gemini to train on your content. Blocking Google-Extended does not remove you from Google AI Overviews (those use the regular Googlebot). For most small shops the calculus favours allowing it, because the citation surface in Gemini's answers is worth more than the training value to Google. - **Q: How big should llms.txt be?** **A:** Aim for 5-30 KB. The ToolGenX version is about 6 KB and 200 lines. Longer files can be parsed but the entire point is to give AI assistants a fast summary, not a full content dump. The detailed pages live at the URLs the file points to. ### My GEO citability scoring sheet: how I grade every page before publishing - **URL:** https://www.agentskillpacks.com/blog/geo-citability-scoring-sheet - **Date Published:** 2026-04-30 - **Tags:** geo-seo, ai-search, content-craft - **Excerpt:** A 100-point checklist for how likely AI search engines are to cite a page. Five categories, the scoring rubric, and the score I will not publish below. - **Quick Answer:** My GEO citability scoring sheet is a 100-point rubric across five categories: answer block quality (30 points), passage self-containment (25), structural extractability (20), authority signals (15), and update freshness (10). I do not publish any page that scores below 75, and in my own tracking the pages above 85 are the ones that turn up in ChatGPT and Perplexity citations. #### FAQs - **Q: Is this scoring sheet the same one used by AI search engines?** **A:** No. I have no insight into what scoring AI search engines actually use. The sheet is reverse-engineered from the published Princeton-GA Tech-IIT Delhi GEO research (2024) plus three months of testing what gets cited versus what does not on my own pages. It is empirical, not insider. - **Q: Why a 75 cutoff and not 80 or 90?** **A:** 75 is the score below which I stopped seeing AI citations on my own pages. Between 75 and 85 they appear occasionally, and above 85 often enough that I treat it as the floor for hub posts. This is a small, uncontrolled sample, so treat the number as a working threshold rather than a finding. - **Q: How long does scoring a page take?** **A:** About 15 minutes the first few times. Five minutes once the rubric is muscle memory. The slow part is reading every passage and asking "could this 60-word block stand alone as an answer to a question". After 50 pages you start to spot the answer-able structure on a quick scan. - **Q: Does the sheet apply to product pages too or just blog posts?** **A:** Both. The categories are the same. Product pages tend to score lower on "update freshness" (they do not get edited often) but higher on "structural extractability" (the JSON-LD does heavy lifting). Blog posts are the opposite. The 75 cutoff applies to both. - **Q: Can a page score 100?** **A:** In theory yes. In practice I have not produced one. The highest I have shipped is 92. The last 8 points tend to require things like multiple original studies cited or first-party data the page is the only source for. Worth knowing they exist as a ceiling but not worth optimizing for. ### Stripe vs Lemon Squeezy vs Paddle vs Gumroad: which checkout for a digital product? - **URL:** https://www.agentskillpacks.com/blog/stripe-vs-iyzico-vs-gumroad-for-digital-products - **Date Published:** 2026-04-28 - **Tags:** payment-infrastructure, solo-founder, stripe, iyzico - **Excerpt:** What each checkout costs on a $29 sale, who files your VAT, how long each one takes to build, and the country rule that decides the question before you do. - **Quick Answer:** Stripe keeps the most of a $29 digital sale at $1.14, then Iyzico at about $1.25, Lemon Squeezy and Paddle at $1.95, and Gumroad at $4.54 on a direct sale. The cheap options leave sales tax and VAT filing to you; the expensive ones are merchants of record and file it for you. Availability decides it first: Stripe onboards businesses only in the countries on its list. #### FAQs - **Q: Which checkout keeps the most of a $29 digital sale?** **A:** Stripe, at $1.14 on a domestic card. Iyzico takes about $1.25, Lemon Squeezy and Paddle $1.95 with tax handled, and Gumroad $4.54 on a direct sale with tax handled. The ranking reverses once you price the hours that VAT registration and filing take, which only the merchant-of-record options remove. - **Q: Is Stripe Tax the same as a merchant of record?** **A:** No, and the difference costs real time. Stripe Tax calculates the tax due and collects it for 0.5% per transaction, but you still register in each jurisdiction and file every return. A merchant of record such as Paddle or Lemon Squeezy becomes the seller of record and files on your behalf, which is what the extra fee buys. - **Q: Is Stripe available everywhere?** **A:** No. A business can open a Stripe account only in the countries on Stripe's global availability list, and that is about where your business is registered rather than where your buyers are. A merchant in a supported country can take payments from cards issued almost anywhere. - **Q: How can a seller in an unsupported country use Stripe?** **A:** The common route is a US company. Stripe Atlas incorporates a Delaware LLC or C corp for a $500 one-time fee plus $100 a year, and that company can open a Stripe account. It also brings US tax filings, so talk to an accountant before treating it as a shortcut. - **Q: What does Iyzico charge per transaction?** **A:** Iyzico's published rate is 4.29% + 0.25 TL per successful transaction, with custom offers for corporate sellers. Installment (taksit) sales carry each bank's own term rate on top. On a $29 sale the standard fee is about $1.25. - **Q: What about Lemon Squeezy and Paddle?** **A:** Both charge 5% + $0.50 per sale with card processing included, and both act as merchant of record, collecting and remitting VAT and sales tax for you. That costs more per sale than Stripe with Stripe Tax, but Stripe Tax only calculates and collects. Registering and filing stay with you. ### Pricing a digital product with Van Westendorp (with real numbers) - **URL:** https://www.agentskillpacks.com/blog/pricing-a-digital-product-with-van-westendorp - **Date Published:** 2026-04-23 - **Tags:** pricing-strategy, solo-founder, business-growth - **Excerpt:** The Van Westendorp Price Sensitivity Meter for solo founders. The four questions, the math, real survey results from a $29 product, and the price I shipped. - **Quick Answer:** The Van Westendorp Price Sensitivity Meter asks four questions about a product (too expensive, expensive but acceptable, cheap but acceptable, too cheap to be credible) and finds the price range where buyers say yes the most. For a $29 digital product I tested, the four-question survey on 87 respondents put the optimal price between $24 and $35, with a clear "indifference point" at $29. I shipped at $29 and the survey called it. #### FAQs - **Q: How many people do I need to survey for Van Westendorp to work?** **A:** Around 50 minimum for a stable result. 100 is comfortable. Below 30 the curves get jagged and the optimal price band can shift by $10 depending on which 10 respondents you happened to ask. I ran 87 and it was enough. - **Q: Where do I find people to survey?** **A:** The same audience that would actually buy. Twitter DMs to followers who asked about pricing, a free download on the existing landing page with the survey as the next page, a small ad budget targeted at the right keywords. Strangers off Mechanical Turk give noisier results because they have no context. - **Q: What if my "too cheap" and "too expensive" curves never cross cleanly?** **A:** Two possibilities. One, your audience is bimodal (two distinct segments with different willingness to pay) and you need to segment before pricing. Two, your product positioning is unclear, so respondents are imagining different products. Fix the positioning, re-run, and the curves cleanly. - **Q: Can I do this for an unreleased product?** **A:** Yes, with caveats. Pre-launch surveys overstate willingness to pay by roughly 20-40% because hypothetical money is easier to spend than real money. A 2005 meta-analysis of 28 studies put the median gap at 35%. Adjust the indifference point down by 25% as a working estimate and validate with the first 10 real sales. - **Q: How often should I re-run Van Westendorp?** **A:** Every 18-24 months, or after a meaningful product change. Markets shift, competitor pricing shifts, your audience matures. The price I shipped at in 2024 is not the price the same survey would suggest in 2026. ### The 6-phase outreach workflow that fills a solo agency pipeline - **URL:** https://www.agentskillpacks.com/blog/6-phase-outreach-workflow-for-a-solo-agency - **Date Published:** 2026-04-16 - **Tags:** business-growth, outreach, solo-founder - **Excerpt:** The cold-outreach system I used to run ModernWebSEO solo. Six phases, real response rates, the email templates that get replies, and research time per lead. - **Quick Answer:** The 6-phase outreach workflow for a solo agency is prospect → research → first-touch email → soft follow-up → proposal → onboarding. Each phase has a fixed time budget per lead (10 min, 15, 5, 5, 30, 60). The system processes 30 leads per week, lands 4-6 meetings, closes 1-2 engagements. Numbers from running ModernWebSEO solo for two years. #### FAQs - **Q: How long should each cold email be?** **A:** Three to five sentences in the first touch. Subject line under 60 characters. No links in the first email. The job of email one is to start a conversation, not to close. Long emails reduce reply rate because the reader has to commit to a wall of text before they know if you are worth it. - **Q: What is a realistic reply rate for cold outreach?** **A:** It depends which activity you mean. Published 2026 benchmarks put average cold email reply rates near 3.1%, with 3-5% described as a realistic target for a volume campaign. Heavily researched outreach at low volume, like the workflow in this post, runs several times higher and cannot be scaled the same way. Compare against the mode you are actually running. - **Q: Do you use sequences or single emails?** **A:** Two-touch maximum. Initial email, one soft follow-up after seven days. Sequences of five or six emails have higher cumulative reply rates but the replies are lower quality and the unsubscribe rate damages domain reputation. Two touches is the cleanest tradeoff. - **Q: What CRM do you use?** **A:** Google Sheets. One row per prospect, columns for the six phase timestamps, a notes field, and a status enum. For a solo agency with 30 leads per week, a spreadsheet beats any CRM tool because the cost of tool maintenance exceeds the value. Switch tools when you have a sales team, not before. - **Q: Should the first email mention pricing?** **A:** No. The first email is to start a conversation. Pricing is the proposal phase. Including price upfront short-circuits the conversation and gets filtered as a sales pitch. The exception is if your pricing is your differentiator (e.g., flat fee in a hourly-billing market). ### GEO + SEO for small shops: what actually moves the needle - **URL:** https://www.agentskillpacks.com/blog/geo-seo-for-small-shops-what-moves-the-needle - **Date Published:** 2026-04-14 - **Tags:** geo-seo, ai-search, solo-founder, structured-data - **Excerpt:** The five SEO + GEO changes that move real traffic for a small shop. No tooling, no agency, no subscription. Just the structural choices that get you cited. - **Quick Answer:** For a small indie shop, five things move real traffic: a passage-quotable writing structure (40-60 word answer blocks under every H2), JSON-LD on every page (Product, Article, FAQPage, Person), a working llms.txt and an AI-crawler-friendly robots.txt, an author Person entity with real sameAs links, and three to ten cornerstone posts that answer your buyer's exact questions. Skip everything else until those five are done. #### FAQs - **Q: Is GEO different from regular SEO?** **A:** They are siblings, not replacements. SEO optimizes for Google's ranking algorithm. GEO optimizes for the extraction layer that pulls passages from your page into ChatGPT, Perplexity, Gemini, and Google AI Overviews answers. The structural work is mostly the same, the emphasis shifts. - **Q: How long does it take to see GEO results?** **A:** Faster than traditional SEO if you have any indexed content. Most AI search engines re-crawl active sites every 24-72 hours, compared to Google's 1-4 weeks for new content. I have seen pages cited in ChatGPT search within a week of publishing. - **Q: Do I need a tool to do GEO?** **A:** For one site with under 50 pages, no. A spreadsheet plus a manual pass through each page with a checklist will get you 80% of the value. Tools become useful when you have 500+ pages or when you want to monitor citation share across competitors. - **Q: Is llms.txt actually used?** **A:** Adoption is mixed but climbing. Anthropic publishes one for its own docs. Google has not committed publicly. The file costs about 10 minutes to write, weighs around 5 KB, and harms nothing. The downside risk is zero, so just ship it. - **Q: My content is good but I do not rank. What is the first thing to check?** **A:** Three things, in order. One, does each page have JSON-LD structured data that matches its type? Two, can a passage from the page stand alone as a 40-60 word answer to a real question? Three, is there a named human author with a real social presence linked from the page? Most small-shop SEO failures live in one of those three. ### Writing for Google AI Overviews: the 7 structural moves that move citations - **URL:** https://www.agentskillpacks.com/blog/writing-for-google-ai-overviews-7-structural-moves - **Date Published:** 2026-04-09 - **Tags:** ai-search, geo-seo, content-craft - **Excerpt:** Seven writing and markup moves that land content in Google AI Overviews, with examples from pages that get cited and pages that do not. - **Quick Answer:** Seven structural moves consistently land content in Google AI Overviews: lead with the answer in 40-60 words, use definition patterns (X is, X refers to), embed one specific number per major section, ship FAQPage schema with real questions, name a human author with Person schema and sameAs, keep paragraph count high and individual paragraphs short, and update dateModified when content meaningfully changes. The moves work because AI Overviews extract passages, not pages. #### FAQs - **Q: How long does it take to appear in AI Overviews after publishing?** **A:** Anywhere from a few hours to two weeks for an already-indexed site. AI Overviews draw on the regular Googlebot index, so the page has to be crawled and indexed first, like any other page. For a brand-new domain, the answer is the same as regular indexation: 1-4 weeks plus authority signals. - **Q: Do AI Overviews care about backlinks?** **A:** Google says AI Overviews use the same index as Search and need no special markup, so authority still counts. What changes is the unit: the answer quotes passages, not pages, so a clear passage from a smaller site can be cited next to bigger ones. - **Q: What is the difference between AI Overviews and Featured Snippets?** **A:** Featured Snippets pull a single passage from a single page and credit that page. AI Overviews synthesize information from multiple pages and cite them as sources. The structural requirements overlap heavily, but AI Overviews rewards being one of N citable sources rather than the one featured source. - **Q: Will my page get cited if it conflicts with the consensus?** **A:** Sometimes. AI Overviews prefer consensus answers but will surface well-structured minority positions for queries where the consensus is weak or contested. Original first-party data or contrarian-but-defensible positions can win citations even against more authoritative sites. - **Q: Does writing for AI Overviews hurt traditional SEO?** **A:** No. The structural moves that work for AI Overviews (clear answers, passage extractability, schema, author entity) are the same moves that improve regular search performance. There is no tradeoff. Writing for one is writing for both. ### The CRO experiment that moved my landing page from 0.4% to 1.9% - **URL:** https://www.agentskillpacks.com/blog/the-cro-experiment-that-moved-04-to-19-percent - **Date Published:** 2026-04-02 - **Tags:** cro, solo-founder, conversion - **Excerpt:** A 30-day rewrite took conversion from 0.4% to 1.9% on 40 conversions. The six changes, the split, the numbers, and what the sample cannot support. - **Quick Answer:** A 30-day split test moved a SaaS template landing page from 0.42% to 1.93% conversion, or 9 purchases to 40. Six changes shipped together: hero rewrite, social proof above the fold, price-anchored CTA, removal of three inherited trust badges, faster LCP via image preload, and an honest what-you-do-not-get section. Follow-up tests put most of the lift on the hero. The sample is small, so read the direction rather than the multiplier. #### FAQs - **Q: How big was the traffic sample for the A/B test?** **A:** About 4,200 sessions over 30 days, roughly 50/50 split between the control and treatment. Below 2,000 sessions per variant the lift was noisy enough that I could not call it statistically significant. 4,000+ produced a 95%+ confidence interval on the result. - **Q: What A/B testing tool did you use?** **A:** A homemade Cloudflare Worker that flipped between two versions of the page based on a hash of the visitor's IP. Free, no third-party JS, no consent banner overhead. For a single-page test this beat Optimizely or VWO because the cost of those tools' tracking JS was itself a measurable LCP hit. - **Q: Could the lift be attributed to other factors (seasonal, referral mix)?** **A:** It is possible. I controlled for referral source by only counting sessions from a specific paid traffic source where the targeting did not change during the experiment. Seasonal effect is unlikely over 30 days for this audience. The conversion lift on the treatment was large enough to be explained by noise at this sample size. - **Q: Which of the six changes did the most work?** **A:** The hero rewrite, by a clear margin. About 60% of the absolute lift came from changing the hero copy from generic SaaS marketing language to a specific concrete answer. The other five changes each contributed 5-15% of the remaining lift. - **Q: Would the same changes work on a different product?** **A:** The pattern would. The specifics would not. The "hero rewrite" was about replacing vague benefit language with a concrete answer-block style, that pattern transfers. The "remove three trust badges" specifically meant the Norton, McAfee and TRUSTe badges I had inherited, and those specific badges hurt more than they helped on this audience. Your trust badges are different. ### One podcast episode to 12 content pieces: the actual repurposing workflow - **URL:** https://www.agentskillpacks.com/blog/one-podcast-to-12-content-pieces-the-actual-workflow - **Date Published:** 2026-03-26 - **Tags:** content-craft, content-multiplication, solo-founder - **Excerpt:** Turn one 45-minute podcast episode into 12 pieces across six platforms over 14 days, with a 4-hour production budget after recording. - **Quick Answer:** The 12-piece content multiplication workflow turns one 45-minute podcast episode into 12 derived pieces: 1 long blog post, 2 Twitter threads, 3 LinkedIn posts, 2 short clips, 1 newsletter section, 3 Reddit answers. Total production time after the episode is recorded is about 4 hours, mostly editorial decisions about which 90-second clips to surface. The bottleneck is the transcript quality, not the writing. #### FAQs - **Q: Do you need a podcast to use this workflow?** **A:** No. Any 30-60 minute structured spoken content works: recorded conversations with friends, voice notes you record while walking, even well-scripted YouTube videos. The "podcast" framing is the most common source format, but the workflow is source-agnostic. - **Q: How long after the episode goes live should the derived pieces ship?** **A:** A 14-day rolling window. Day 0: episode + blog post + LinkedIn post 1. Day 2: Twitter thread 1. Day 4: short clip 1. Day 7: LinkedIn post 2, Reddit answer 1. Day 10: thread 2, short clip 2. Day 14: newsletter section. The cadence avoids the firehose-then-silence problem. - **Q: What is the transcript quality cutoff for this to work?** **A:** Word error rate under 8% is the practical minimum. Above 8% you spend more time fixing transcript errors than writing derived pieces. Whisper large-v3 on clean audio sits around 3-5%; cheap transcription services sit around 10-15%. Pay for the good transcription or record clean audio. - **Q: How much of the workflow can be automated?** **A:** About 60%. Transcription, clip identification, thread drafting, and newsletter excerpt extraction are automatable. Editorial selection (which 90 seconds to clip, which thread to lead with, which Reddit subreddit makes sense) is judgment work that does not automate cleanly. - **Q: What if my podcast is interview-style, not solo?** **A:** Interview content actually produces more derived pieces per episode because each guest's framing is a fresh perspective. The workflow is the same; the LinkedIn posts can sometimes be split between your view and the guest's quoted view, which doubles the surface area. ### Why I rewrote my 19-product site in two weeks (and what I would do differently) - **URL:** https://www.agentskillpacks.com/blog/why-i-rewrote-my-19-product-site-in-two-weeks - **Date Published:** 2026-03-22 - **Tags:** solo-founder, postmortem, site-architecture - **Excerpt:** An honest postmortem of a site rebuild: why the old version failed despite looking fine, what the rewrite changed, and the four decisions I made too late. - **Quick Answer:** I rewrote my 19-product digital shop in three weeks (the title rounds down) because the old site looked fine and had three structural problems no amount of polish could fix: a checkout redirect that hid the funnel, generic AI-template copy that differentiated nothing, and no AI search surface at all. The rewrite fixed the copy and the search surface. The lesson is that a polished site cannot rescue a wrong premise, and looking fine is not evidence. #### FAQs - **Q: Why three weeks and not faster?** **A:** Two weeks of code, one week of content. The code is the easy part once you have done it a few times. The content rewrite, meaning 19 product descriptions through humanizer plus four blog posts, took as long as the entire technical build. - **Q: Did the rewrite immediately improve conversion?** **A:** I do not know yet. The rewrite shipped two weeks ago. I am writing this so future-me has a reference for what we changed and why. Real conversion data will come in over the next 90 days. - **Q: Would you rewrite again from scratch or refactor?** **A:** Refactor, but only if the foundation is sound. The old ToolGenX had unfixable foundation problems: a redirect baked into the buy flow, generic product copy, no GEO surface. None of that is a refactor. A rewrite was the right call. - **Q: How do you avoid scope creep on a rewrite?** **A:** Write the PRD before touching code. Be specific about what is in scope and what is out. Mine had a "kapsam dışı" (out of scope) section that saved me three times when I was tempted to add a feature mid-build. - **Q: What is the one thing you would do differently?** **A:** Write the cornerstone blog posts first, then build the site. Doing it the other way meant I had to rewrite product copy after the fact to match a voice I had not yet figured out. Cost me roughly a week of extra editing. ## 5. Topic Archives Every blog tag has its own archive page. The count is how many posts carry that tag. - **solo-founder** (18 posts): https://www.agentskillpacks.com/blog/tag/solo-founder - **ai-search** (7 posts): https://www.agentskillpacks.com/blog/tag/ai-search - **ai-workflow** (7 posts): https://www.agentskillpacks.com/blog/tag/ai-workflow - **claude-code** (6 posts): https://www.agentskillpacks.com/blog/tag/claude-code - **geo-seo** (6 posts): https://www.agentskillpacks.com/blog/tag/geo-seo - **business-growth** (5 posts): https://www.agentskillpacks.com/blog/tag/business-growth - **content-craft** (5 posts): https://www.agentskillpacks.com/blog/tag/content-craft - **developer-tools** (5 posts): https://www.agentskillpacks.com/blog/tag/developer-tools - **payment-infrastructure** (5 posts): https://www.agentskillpacks.com/blog/tag/payment-infrastructure - **pricing-strategy** (4 posts): https://www.agentskillpacks.com/blog/tag/pricing-strategy - **postmortem** (2 posts): https://www.agentskillpacks.com/blog/tag/postmortem - **seo** (2 posts): https://www.agentskillpacks.com/blog/tag/seo - **content-and-writing** (1 post): https://www.agentskillpacks.com/blog/tag/content-and-writing - **content-multiplication** (1 post): https://www.agentskillpacks.com/blog/tag/content-multiplication - **conversion** (1 post): https://www.agentskillpacks.com/blog/tag/conversion - **cro** (1 post): https://www.agentskillpacks.com/blog/tag/cro - **digital-products** (1 post): https://www.agentskillpacks.com/blog/tag/digital-products - **eu-compliance** (1 post): https://www.agentskillpacks.com/blog/tag/eu-compliance - **geo** (1 post): https://www.agentskillpacks.com/blog/tag/geo - **humanizer** (1 post): https://www.agentskillpacks.com/blog/tag/humanizer - **iyzico** (1 post): https://www.agentskillpacks.com/blog/tag/iyzico - **legal** (1 post): https://www.agentskillpacks.com/blog/tag/legal - **llms-txt** (1 post): https://www.agentskillpacks.com/blog/tag/llms-txt - **outreach** (1 post): https://www.agentskillpacks.com/blog/tag/outreach - **site-architecture** (1 post): https://www.agentskillpacks.com/blog/tag/site-architecture - **stripe** (1 post): https://www.agentskillpacks.com/blog/tag/stripe - **structured-data** (1 post): https://www.agentskillpacks.com/blog/tag/structured-data ## 6. Site Map and Machine-Readable Endpoints - **HTML sitemap:** https://www.agentskillpacks.com/sitemap - every public page grouped and annotated with a one-line description. - **XML sitemap:** https://www.agentskillpacks.com/sitemap.xml - **Crawl rules:** https://www.agentskillpacks.com/robots.txt - **Curated model map:** https://www.agentskillpacks.com/llms.txt - **Products feed:** https://www.agentskillpacks.com/feeds/products.xml - **Blog feed:** https://www.agentskillpacks.com/feeds/blog.xml - **Books feed:** https://www.agentskillpacks.com/feeds/books.xml - **Markdown mirror:** send `Accept: text/markdown` to any product, category, book or blog URL to get the source text instead of HTML.