Tom Witt
Community@tomw200082-collab
Tom Witt's registry combines Claude Code engineering discipline skills with a Hebrew-language GT Factory OS suite covering production planning, procurement, Shopify sync, and delivery logistics.
Agent Skills by Tom Witt
Showing 50 vetted skills indexed across 1 GitHub repositories.
tomw200082-collabusing-git-worktrees
Creates isolated git worktree workspaces with detection, setup, and baseline test verification.
tomw200082-collabcaveman-help
Displays a quick-reference card of caveman modes, skills, and configuration options.
tomw200082-collabstop-slop
Removes AI writing patterns from prose during drafting, editing, and review.
tomw200082-collabtest-driven-development
Enforces red-green-refactor test-driven development workflow before writing implementation code.
tomw200082-collabsystematic-debugging
Diagnose bugs through a four-phase root cause investigation process before proposing fixes.
tomw200082-collabcopywriting
Write and improve marketing copy for homepages, landing pages, pricing pages, and CTAs.
tomw200082-collabcaveman-review
Generates one-line code review comments with location, severity, problem, and fix.
tomw200082-collabmasterprompt
Writes self-contained handoff briefs for executors lacking the original conversation context.
tomw200082-collabcaveman-compress
Compress natural language memory files into terse caveman format to reduce input tokens.
tomw200082-collabusing-superpowers
Enforces skill discovery and invocation before any agent response or action.
tomw200082-collabdispatching-parallel-agents
Dispatches concurrent subagents to fix independent test failures across separate problem domains.
tomw200082-collabclose-session
Detects open loops, runs safe cleanup, and preserves session knowledge at work-session close.
tomw200082-collabcopy-editing
Reviews and improves marketing copy through seven focused editing passes.
tomw200082-collabponytail
Enforces minimal code solutions by prioritizing stdlib, native features, and existing code over new dependencies.
tomw200082-collabexecuting-plans
Executes written implementation plans with review checkpoints and task verification.
tomw200082-collabfinishing-a-development-branch
Guides merging, pushing, or preserving a completed Git branch with test verification and worktree cleanup.
tomw200082-collabponytail-audit
Audits an entire codebase for over-engineering and ranks findings to delete or simplify.
tomw200082-collabdaily-delivery-dispatch
Assigns in-stock LionWheel delivery orders to a driver and date after stock verification.
tomw200082-collabponytail-help
Displays a quick-reference card for ponytail modes, skills, and commands.
tomw200082-collabskill-creator
Create, evaluate, and iteratively improve Claude skills with benchmarked test runs.
tomw200082-collabroute-print-pack
Builds a print-ready daily delivery route pack PDF from LionWheel and Green Invoice data.
tomw200082-collabcavecrew
Guides delegation to compressed-output subagents for code location, editing, and review tasks.
tomw200082-collabapple-design
Review mobile and desktop UI designs against Apple HIG-derived design guidelines.
tomw200082-collabbrainstorming
Guides collaborative design dialogue to turn ideas into approved specs before implementation.
Frequently Asked Questions About Tom Witt
FAQPage SchemaWhat tasks can I automate using Tom Witt's skills?▼
The skills cover two territories: Claude Code engineering practices (TDD, systematic debugging, code review, git worktrees, plan execution) and GT Everyday business operations including production planning, procurement sessions, Shopify inventory sync, goods receipt from supplier invoices, delivery dispatch, and Hebrew sales reports.
Who are these skills designed for?▼
Primarily Tom operating GT Everyday's factory and e-commerce business in Hebrew, plus software engineers using Claude Code who want disciplined development practices like brainstorming-before-building, subagent-driven execution, over-engineering audits (ponytail), and token-compressed communication (caveman).
How do the GT Factory OS skills run in practice?▼
They fire on scheduled triggers (06:30 daily guardian, Wednesday production planning, 20:00 evening run) or Hebrew/English voice commands. They read live systems like Shopify, Notion, and LionWheel, draft plans and purchase orders, and always pause for Tom's approval before any write to external systems.
Are these skills open source and what do they cost?▼
Several skills carry explicit MIT licenses (stop-slop, ponytail, ogilvy); others have no license declared. All are prompt-based skills with no stated fees, though business skills depend on live GT systems like Shopify, LionWheel, Make.com webhooks, and Notion.
What prerequisites do these skills require?▼
Engineering skills need a Claude Code session with git. Business skills require GT Factory OS infrastructure: Shopify store access, LionWheel account, Notion as task master, drivers.json and route_calendar.json configs, Green Invoice, and a Make.com-to-Gmail webhook for branded Hebrew email reports.