create-ultimate-skill_skill

This skill guides you through creating, reviewing, and iterating Claude Code skills using official practices and up-to-date API docs.
  • Python

30

GitHub Stars

1

Bundled Files

2 months ago

Catalog Refreshed

4 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill henkisdabro/wookstar-claude-plugins --skill create-ultimate-skill

  • SKILL.md9.2 KB

Overview

This skill creates, reviews, and iterates on Claude Code skills using official best practices and the latest API constraints. It guides definition of use cases, archetype selection, frontmatter composition, folder layout, and validation scripts. The skill also offers a structured review mode to find critical, major, and minor issues and produce prioritized fixes.

How this skill works

It asks targeted questions to capture 2–3 concrete use cases, negative triggers, archetype, triggers, and scope. It fetches the latest official documentation to validate frontmatter fields and constraints, scaffolds files and utility scripts, and runs automated checks to enforce naming, description, and structure rules. In review mode it parses the skill package, categorizes issues by severity, and generates a concise remediation plan.

When to use it

  • When asked to create a new Claude Code skill or when the user says "create a skill"
  • When reviewing an existing skill and wanting a structured audit or prioritized fixes
  • When writing or improving a skill description, triggers, or frontmatter to meet constraints
  • When designing skill architecture, folder layout, or validation scripts
  • When iterating on triggering behaviour to reduce false positives or missed activations

Best practices

  • Collect 2–3 concrete positive use cases and explicit negative triggers before drafting the skill
  • Keep the description a single-line plain string, third-person, and include trigger phrases
  • Enforce name constraints: lowercase, hyphens, ≤64 chars, only a-z0-9-
  • Limit references to one level deep and use forward slashes only for paths
  • Provide deterministic scripts for validation and concrete examples for functional tests
  • Define quantitative and qualitative success criteria and iterate on the hardest case first

Example use cases

  • Create a methodology-style skill that automates a repeatable workflow with the exact prompt and examples
  • Review a skill bundle to find critical frontmatter issues, missing triggers, and oversized documentation
  • Convert an ad-hoc process into a CLI Reference archetype with grouped commands and minimal prose
  • Package a validated skill into a distributable zip while checking naming and frontmatter rules
  • Tune triggers and negative triggers when a skill activates too often or never activates

FAQ

Multi-line descriptions or YAML lists for allowed-tools break frontmatter parsing; always use a single-line plain string and comma-separated allowed-tools.

How do I check trigger quality?

Define success metrics, test direct and paraphrased triggers plus edge cases, and iterate by adding or tightening trigger phrases.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational