skill-management_skill

This skill guides you through creating, updating, and maintaining Claude Code skills, ensuring best practices and committed changes to the repository.
  • Python

1

GitHub Stars

2

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 dbosk/claude-skills --skill skill-management

  • LICENSE.md1.0 KB
  • SKILL.md16.7 KB

Overview

This skill codifies a safe, repeatable workflow for creating, updating, and validating Claude Code skills under ~/.claude/skills/. Activate this skill before any edit to a skill’s files. It enforces structure, progressive disclosure, bundling patterns, and an obligatory git commit step after changes. Use it to avoid rework and maintain high-quality, reusable skills.

How this skill works

Inspect the target skill directory and ensure required frontmatter and manifest conventions are present. Guide the author through planning reusable resources (scripts, references, assets), initializing the skill structure, implementing bundled resources, validating with provided scripts, testing triggers, and committing changes to the skills git repo. Provide a short checklist and corrective actions for common validation errors.

When to use it

  • Before creating a new skill in ~/.claude/skills/
  • Before editing any main skill manifest or bundled files
  • When asked to review, improve, refactor, or test a skill
  • When defining triggers, bundled scripts, references, or assets
  • Whenever you need to validate or test skill loading and behavior

Best practices

  • Write a concise manifest that states what the skill does and when to use it
  • Use progressive disclosure: keep the main file small and put long material in references/
  • Include scripts/ for deterministic operations and assets/ for output templates
  • Adopt imperative writing (verb-first) and provide defaults for options
  • Run the included validation script and fix reported issues before committing
  • Commit logical, documented changes to the skills git repo after finishing edits

Example use cases

  • Create a new image-editor skill: plan assets, add scripts for repeatable transforms, then initialize and validate
  • Update a skills trigger set: review existing manifests, adjust trigger phrases, test invocation, then validate
  • Add a reference file with domain schemas and link it for progressive loading
  • Refactor brittle scripting into scripts/ so behaviors are deterministic and easy to test
  • Run quick validation on a modified skill directory and resolve lint or frontmatter errors

FAQ

Include YAML frontmatter with name and description that states what the skill does and when to use it.

When should I add scripts/ versus writing instructions inline?

Add scripts/ when behavior must be deterministic or the same code is reused frequently; keep long explanations in references/ instead of inline.

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