From cloudbase-ai-toolkit
Creates and refines SKILL.md files with better trigger descriptions, progressive disclosure, and testable behavior. Audits collections for redundancy and evaluates trigger quality.
How this skill is triggered — by the user, by Claude, or both
Slash command
/cloudbase-ai-toolkit:skill-authoringThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Create and refine reusable agent skills with better trigger quality, cleaner structure, stronger behavioral guidance, and more reliable evaluation.
Create and refine reusable agent skills with better trigger quality, cleaner structure, stronger behavioral guidance, and more reliable evaluation.
Use this skill when you need to:
SKILL.mdname or descriptionconfig/source/skills for redundancy, trigger overlap, or weak boundariesSKILL.md plus references/, assets/, or scripts/When the task targets config/source/skills, apply these guardrails in addition to the normal skill-authoring workflow:
This section is the repo-managed CloudBase skill review baseline for this repository.
versionDo NOT use for:
SKILL.mdIdentify the task class first
Optimize the trigger surface early
name and especially description before expanding the bodydescription, not only into the bodyDesign behavior, not just documentation
SKILL.md tell the agent what to do after the skill triggersLoad supporting materials only when needed
Use collection-level review when the request is about many skills
config/source/skills, check overlap, duplication, trigger boundaries, and progressive disclosure across neighboring skillsEvaluate before considering the skill complete
| Task | Read |
|---|---|
Write or improve name and description | references/frontmatter-patterns.md |
| Design skill anatomy and progressive disclosure | references/structure-patterns.md |
| Draft a new skill or review an existing one | references/templates.md |
Audit config/source/skills for quality, redundancy, and overlap | references/repo-skill-review.md |
| Review repo-managed CloudBase source skills | references/cloudbase-skill-review.md |
| Build evaluation prompts and review outcomes | references/evaluation.md |
| Compare good examples, weak examples, and rewrites | references/examples.md |
name and description with realistic trigger language.config/source/skills, read references/repo-skill-review.md, then load references/cloudbase-skill-review.md for CloudBase-specific standards before proposing rewrites.SKILL.md so it changes agent behavior after trigger.references/, assets/, or scripts/ as needed.name short, intentional, and stable?description explain both capability and trigger conditions?SKILL.md change agent behavior after trigger?npx claudepluginhub tencentcloudbase/cloudbase-mcpGuides authoring and validating Claude Code skills: metadata optimization, lean workflows, progressive disclosure, references, skillgrade evals.
Creates, evaluates, and reviews Claude Code skills and agent definition files. Covers skill-vs-hook decisions, SKILL.md authoring, frontmatter optimization, baseline-first evals with MCP tooling, validation, and packaging.
Creates, fixes, analyzes, and validates skills for AI agents. Includes scripts for structure checks, quality analysis, and character budget management.