Wizalnut

Most spells don'twork. This one does.

We run a video studio on Claude Code. Every hook, skill and agent that stops our agents from guessing, leaking and forgetting is becoming four packs. The first one is out. Install in ten minutes.

The same titles
run our ads.

Built in After Effects for our own campaigns. Nothing on this page is a mockup.

Four pages
from the book.

A spell gets into the book only after it worked. These four did, on paid client work.

PAGE 01

Quality Gates

Hooks that stop Claude Code before it guesses, builds on stale facts, leaks a key or calls unfinished work done.

  • Done gate: no “done” without a real check after the last edit
  • Premise check: claims from old notes are verified against live files first
  • Git guards: no discarding uncommitted work, no blind amend
  • Secret guards: keys never echoed into logs, traces or prompts
  • 16 hooks, 6 skills, a canary test for every hook, one-command install and uninstall
€29Buy now

PAGE 02

Memory Architecture

A memory your agent reads in the right order, with a guard that flags every note nobody can find again.

  • Router read every session, indexes read only when needed
  • Index guard: an unlinked note is flagged as it is written
  • Decision log: exact words first, interpretation second
  • Handoff template, so the next session starts where this one ended
  • Weekly hygiene pass for duplicates and stale notes
€29In packaging

PAGE 03

Agent Team

Twenty senior specialists, a dispatcher that picks the right one, and a contract for how they hand work over.

  • QA, security, DevOps, docs, finance, legal, marketing, growth
  • Dispatcher that routes a vague task to the right specialist
  • Orchestrator that turns a decision into today's plan
  • Handoff contract: agents pass on work, not opinions
  • Routing scale: when to answer alone, when to delegate
€39In packaging

PAGE 04

AI Influencer Kit

One anchor image, and the same face in every angle, outfit, place and talking head. A photoreal person and a painted mascot.

  • One anchor for angles, expressions, outfits, poses, locations
  • Double pack: photoreal person and painted mascot
  • 70+ tested prompts and a copy-paste file
  • Talking heads: front camera, tight 135 mm side, wider three-quarter
  • Location lock, so the place does not redraw itself
€39In production

What the book
caught today.

Three gates from the Quality Gates pack, as they fired in our own working sessions on 11 October 2026. Real output, shortened and translated from Slovak.

Blocked “done” without proof

“Done-claim detected, but there was a code edit after the last check. Run a real test, or rewrite the message without saying done.”

The agent rendered the page and checked it before calling it finished.

Blocked discarding work

“git checkout would discard uncommitted changes. In a shared tree they may belong to another session. Look at the diff first.”

The agent proved the change was its own before touching the file.

Stopped stale facts

“First build of this session. Does it rest on claims from old notes or a handoff? Verify them against the live files before you continue.”

Claims from the morning notes were checked against live files before any file was written.

It's a page from my book. The page is yours.

All four packs, with every update we ship. The same files that run our studio. Opens when all four are ready.

Quality GatesMemoryAgent TeamInfluencer Kit

€99€136

Before
you ask.

When can I buy?

Quality Gates is available now. The other three open as soon as each one is finished. Follow @wizalnut on Instagram and you will hear the day they do.

Is this what you use yourselves?

Yes. The same hooks, skills and agents that run our client work, with client names and private paths removed and the text translated to English.

How do I install it?

Unzip and run bash install.sh. It backs up your settings.json, registers the hooks and installs the skills; --dry-run shows every change first and --uninstall removes it all. Then start a new Claude Code session.

What do I need?

Claude Code 2.x, bash, python3 and jq. Built and tested on macOS; hooks are shell scripts and skills are Markdown files.

Will it slow my agent down?

Most hooks are a quick pattern check that stays silent. They interrupt only when a rule is about to be broken, and every message says exactly what to do next.

Can I change the rules?

Every hook is a readable script with its reason written at the top, and each one has an off switch.