AI Tool ConsultantClaude writes the format, not the lessons
Ask Claude for a skill and you get its default approach, tidied up. That's what it would have done anyway, so nothing changes.
27 skills for Claude, Claude Code and Codex
AI skills built from real projects.
Each skill installs a complete workflow, plus the checks people usually learn the hard way: the spreadsheet tab nobody read, the number nobody re-derived, the internal file a website kept serving. Install once, then just ask.
One-time purchaseFiles you keepYour data stays in your own AI tools
example-run.log
claude "do the monthly LinkedIn report"
› using skill: linkedin-analytics-report
✓ inventoried 3 workbooks, every sheet listed
✓ followers AND visitors compared, not just followers
✓ competitor identities checked on their own sites
! forecasts disagree: 28 (recent pace) vs 51 (campaign pace)
→ showing both scenarios, not one number
✓ 10 slides rendered, every slide looked at
Works with
Claude.aiClaude DesktopClaude CodeCodexany agent that reads SKILL.md
coding-agent-handovercoding-agent-session-startsocial-media-caption-writerexam-prep-tutorlinkedin-ghostwriterlinkedin-analytics-reportlinkedin-carousel-makercold-outreach-sequence-writerlemlist-campaign-auditkeyword-researchseo-content-strategytechnical-seo-auditseo-blog-writerblog-content-refreshgerman-writing-correctiongerman-learning-coachlanguage-speaking-coachbilingual-karaoke-subtitlesllm-evaluation-integrityllm-output-validationfact-check-before-sendingmulti-agent-repo-setupovernight-agent-reportmanual-acceptance-checklistllm-feature-guardrailsdeploy-exposure-checkbuild-debug-practice-modescoding-agent-handovercoding-agent-session-startsocial-media-caption-writerexam-prep-tutorlinkedin-ghostwriterlinkedin-analytics-reportlinkedin-carousel-makercold-outreach-sequence-writerlemlist-campaign-auditkeyword-researchseo-content-strategytechnical-seo-auditseo-blog-writerblog-content-refreshgerman-writing-correctiongerman-learning-coachlanguage-speaking-coachbilingual-karaoke-subtitlesllm-evaluation-integrityllm-output-validationfact-check-before-sendingmulti-agent-repo-setupovernight-agent-reportmanual-acceptance-checklistllm-feature-guardrailsdeploy-exposure-checkbuild-debug-practice-modes
27skills in 6 packs, 4 of them free
55+evaluation runs behind the LLM pack, zero retracted numbers
22live outbound campaigns audited for the Lemlist skill
35/49reviews where a model broke its one rule, caught by a check in code
The honest question
Why buy a skill when Claude can write one?
Because the part that matters is the part Claude doesn't know yet.
01
02
The value is in what went wrong
Every rule here exists because a real run failed: a sheet nobody read reversed a conclusion, a "better" score came from server drift, a private repo's site served internal plans.
03
You can't ask for a check you don't know you need
That's what you're buying: failures someone already had, written as instructions, scripts and templates your AI follows every time.
Packs
Pick the workflow you repeat every week.
One-time price per pack. All 27 skills for €129. Checkout opens soon.
For founders & B2B marketers
LinkedIn Ghostwriting & Outreach
Write in a real person's voice, report what LinkedIn actually did, and run outbound that doesn't burn your domain.
5 skills
€49
For content & SEO teams
SEO & Blog Content
Keyword research from evidence, one URL per keyword, and articles where every claim has a source.
5 skills
€49
For German learners
Learn German with AI
The daily method behind months of measured German progress, as skills Claude can run for you.
4 skills
€29
For AI engineers & researchers
LLM Evaluation & Fact-Checking
Report AI results you never have to retract.
3 skills
€39
For builders using coding agents
Claude Code & Codex Workflow
Run two coding agents on real products without losing the thread, or your secrets.
6 skills
€29
Free
Start with four skills, free.
The same standard as the paid packs, so you can judge the quality before you buy anything.
Free
Coding Agent Handover
End every AI coding session with a handover the next session can act on cold.
No setup neededView →
Free
Coding Agent Session Start
A two-minute preflight so the agent works in the right repo, branch and scope.
gitView →
Free
Social Media Caption Writer
Platform-native captions for Instagram, TikTok, YouTube Shorts and LinkedIn, checked against current limits.
Python 3 (for the checker)View →
Free
Exam Prep Tutor
Exercises first, honest scores, and a record that never invents progress.
No setup neededView →
How it works
Install once. Then just ask.
A skill switches on when your request matches it. "Do the monthly LinkedIn report" is enough.
STEP 1
Choose a pack
Or start with the free kit. Every skill page says exactly what it does and what it needs.
STEP 2
Install it
A Claude Code plugin, a zip for Claude.ai, or a folder for Codex. Two minutes.
STEP 3
Ask normally
The skill brings its workflow, scripts and checks. You review the result.
# Claude Code
/plugin marketplace add <your-invited-repo>
/plugin install <pack>@aitoolconsultant
# Claude.ai / Desktop: Customize → Skills → + → Upload a skill (one zip per skill)
# Codex: copy the skill folders into ~/.agents/skills/
linkedin-analytics-report/
├── SKILL.md # the workflow + the checks
├── scripts/
│ ├── inventory_workbooks.py # reads every sheet
│ └── parse_visitor_list.py # proves the count
├── references/
│ ├── pitfalls.md # 7 real incidents
│ └── deck-style.md
└── agents/openai.yaml # Codex metadata
How these were built
Strategy → task → execution → validation → handover.
The operating system behind every skill here, and behind the products they came from. Read the field note →
01 · strategy
Think before building
Product reasoning and task design happen in a separate session from the code.
02 · task
Scope one slice
One reviewable change with explicit non-goals and files to inspect.
03 · execute
Use repo-aware agents
Claude Code or Codex make targeted edits and run the checks.
04 · validate
Smoke before ship
Real browser, nonsense input, cheap guards before expensive model calls.
05 · handover
Write the memory
Files, checks, decisions, risks and the next step, so the next session starts clean.
Built by
I'm Aviral. I build with AI agents and write down what breaks.
I'm a founder and operator. I build AI products (myberuf, a German-for-work practice app), work on growth and AI operations at a B2B start-up, and ran a local-LLM evaluation project during my Master's. These skills are the checklists I wish I'd had at the start: one builder's operating desk, not an agency.
From the blog
The long version, for free.
Guides to Claude Skills, coding agents and AI search, written from real builds. All articles →
Claude Skills
What Are Claude Skills? How Agent Skills Work (2026 Guide)
Claude Skills are folders of instructions, scripts and checks that Claude loads when a task needs them. How they work, where they run, which to use.
Claude Code & Codex
The Claude Code Handover: End Every Session With a File the Next One Can Act On
A Claude Code handover is the file a fresh session reads first. The template, the rules behind it, and a ready-to-paste prompt for the next chat.
Claude Skills
SKILL.md Format Explained: Frontmatter, Folder Structure and Examples
SKILL.md field by field: frontmatter rules, extra fields Claude Code accepts and Claude.ai rejects, folder structure, limits and a full example.
FAQ
Questions, answered plainly.
When can I buy?
Soon: checkout isn't open yet. Follow on LinkedIn to hear when it opens.
Why buy a skill when I could ask Claude to write one?
Claude can write the format of a skill in a minute, but only from what it already knows. A skill generated from a prompt contains Claude's default approach, the same thing it would have done without the skill, so nothing changes.
These skills are worth something because they override those defaults with checks that came from real failures: the spreadsheet tab nobody read, the improvement that was really server drift, the internal file a website kept serving. You can't ask for a check you don't know you need. That's what you're buying.
Which tools do the skills work with?
Claude.ai and Claude Desktop (upload a zip), Claude Code (plugin or folder), Codex (folder), and other agents that read the open SKILL.md format. A few skills include Python scripts; each skill page lists exactly what it needs.
How do I install them?
In Claude Code, add the private marketplace you're invited to after purchase and install the pack with one command. For Claude.ai and Claude Desktop, turn on code execution in Settings → Capabilities, then upload one zip per skill under Customize → Skills. For Codex, copy the skill folders into ~/.agents/skills/. Then ask normally; a skill switches on when your request matches it.
Do I get updates?
Yes. In Claude Code, turn on auto-update for the marketplace (/plugin → Marketplaces), or update a pack with claude plugin update <pack>@aitoolconsultant. Zip buyers can download the latest version from their purchase page.
Can I use them for my team or my clients?
Yes, for your own work and for client work you do. You can share a pack inside one team of up to five people at one company. Please don't resell or publish the files.
Are these official Anthropic or OpenAI products?
No. They're independent skills that follow the open Agent Skills (SKILL.md) format, built and maintained by one person.
Do you see my data?
No. Skills are instructions and scripts that run inside your own AI tool. Nothing is sent to us.
Four skills. No cost. See the difference.
Handover, session start, social captions and an honest exam-prep tutor, free in exchange for your email.