Agent skill
wizardshit
Work on wizardshit.ai — Joel's AI education platform teaching the exploratory journey of building systems with and around AI. Covers the Next.js app in gremlin, The Claw course design, and the manage brain vault. Triggers on: 'wizardshit', 'wizard shit', 'wizardshit.ai', 'the claw course', 'agent game generation', 'zots', 'minions of toil'.
Install this agent skill to your Project
npx add-skill https://github.com/joelhooks/joelclaw/tree/main/skills/wizardshit
SKILL.md
wizardshit.ai
Joel's AI education platform. The crux: teach the exploratory journey of building systems with and around AI.
Not a polished curriculum factory — a system that teaches by showing the system being built. Digital garden philosophy applied to AI education.
Repositories
App: ~/Code/badass-courses/gremlin/apps/wizardshit-ai/
Next.js 16 app in the gremlin monorepo (badass-courses/gremlin). Uses the badass-courses platform:
@badass-courses/core,@badass-courses/sdk,@badass-courses/next,@badass-courses/ui- Convex backend via
@badass-courses/convex-adapter - Package name:
@wizardshit/web
The gremlin monorepo has its own AGENTS.md — read it before any gremlin work:
cat ~/Code/badass-courses/gremlin/AGENTS.md
Brain/Design: ~/Code/joelhooks/manage/
Vercel manage repo with PARA-structured brain vault. Course design docs live here:
brain/01 Projects/Agent Game Generation/— The Claw course design bibleskills/the-claw-course-design/SKILL.md— course design process skilldocs/decisions/— ADRs for the manage workspace
Key design documents (read in order):
Course Concept.md— canonical spec (~5K words)Lesson Bookends.md— first and last lesson, fully workshoppedGame Design Patterns.md— 20 patterns mapping game mechanics → agent architectureResearch - Schell Lenses x Dwarf Fortress x The Claw.mdResearch - 4CID Learning Design Bible.mdResearch - Dwarf Fortress Design Philosophy.mdResearch - just-bash as Game Engine.mdResearch - Game Design Bibles.md— Koster + SchellResearch - Prior Art and Landscape.mdResearch - Multi-Agent Frameworks as MUD Architecture.mdResearch - MUD Architecture and Platform Primitives.md
Public gist (full design docs):
https://gist.github.com/joelhooks/99b7188fdd775b32b646ee72484774cc
Two Versions
- Vercel Academy — toned-down business version. Vercel-native stack (Workflow, Sandbox, Queues, AI Gateway, just-bash). Draft: https://vercel.com/academy/draft/agent-game-generation
- wizardshit.ai — the full unfiltered version. Free to use whatever infra. Inngest, k8s, Redis, self-hosted — whatever's best.
The Claw: Minions of Toil
The flagship course. A roguelite colony management sim where Zots (autonomous AI agents) toil in a fantasy dungeon generated from real codebases.
Core insight: the game mechanics ARE agent architecture patterns. Health bar = context window. Abilities = tools. Party = multi-agent delegation. Combat = error handling. The representation IS the thing.
Design foundations:
- Koster — fun = learning patterns in a safe space
- Schell — 100+ lenses for design evaluation
- 4C/ID — whole-task from day 1, scaffolding fades, driven by what breaks
- Dwarf Fortress — colony management, autonomous actors, "losing is fun", emergent narrative
Tech (wizardshit version — free to choose):
- just-bash — the game engine (virtual filesystem sandbox)
- AI SDK — Zot brains
- Inngest — tick engine, durable workflows
- Whatever persistence/state makes sense
Linear: DX-2875
Working With This Project
- For gremlin app work:
cd ~/Code/badass-courses/gremlinand read its AGENTS.md - For course design:
cd ~/Code/joelhooks/manageand read the brain vault docs - For joelclaw integration: this is where wizardshit infra decisions may overlap with joelclaw patterns
- Load the
gremlinskill for gremlin-repo-specific guidance
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
joelhooks/joelclaw
skill-review
Audit and maintain the joelclaw skill inventory. Use when checking skill health, fixing broken symlinks, finding stale skills, or running the skill garden. Triggers: 'skill audit', 'check skills', 'stale skills', 'skill health', 'skill garden', 'broken skill', 'skill review', 'fix skills', 'garden skills', or any task involving skill inventory maintenance.
joelhooks/joelclaw
cli-design
Design and build agent-first CLIs with HATEOAS JSON responses, context-protecting output, and self-documenting command trees. Use when creating new CLI tools, adding commands to existing CLIs (joelclaw, slog), or reviewing CLI design for agent-friendliness. Triggers on 'build a CLI', 'add a command', 'CLI design', 'agent-friendly output', or any task involving command-line tool creation.
joelhooks/joelclaw
daily-summary
Generate a daily system activity summary across all joelclaw subsystems. Triggers on 'daily summary', 'what happened today', 'system report', 'today's tally', 'activity summary', 'what did we do', 'daily report'.
joelhooks/joelclaw
inngest-durable-functions
Create and configure Inngest durable functions. Covers triggers (events, cron, invoke), step execution and memoization, idempotency, cancellation, error handling, retries, logging, and observability.
joelhooks/joelclaw
recall
Fan-out search across all memory sources when context is unclear or vaguely referenced. Triggers on: 'from earlier', 'remember when', 'what we discussed', 'that thing with', 'the conversation about', 'did we ever', 'what happened with', 'you mentioned', 'we talked about', 'earlier today', 'last session', 'the other day', or any vague reference to past context that needs resolution before the agent can act.
joelhooks/joelclaw
inngest-local
Set up self-hosted Inngest on macOS as a durable background task manager for AI agents. Interactive Q&A to match intent — from Docker one-liner to full k8s deployment with persistent state. Use when: 'set up inngest', 'background tasks', 'durable workflows', 'self-host inngest', 'event-driven functions', 'cron jobs', or any request for a local workflow engine.
Didn't find tool you were looking for?