agent-creator

Create high-quality Claude Code agents from scratch or by adapting existing agents as templates. Use when the user wants to create a new agent, modify agent configurations, build specialized subagents, or design agent architectures. Guides through requirements gathering, template selection, and agent file generation following Anthropic best practices (v2.1.63+).

Quality

81%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Quality

Content

70%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

Highly actionable with a well-structured, validated workflow, but the body is padded with pedagogical sections that re-explain concepts Claude already knows and inlines material that belongs in the reference files it already ships.

Suggestions

Move 'Common Mistakes', 'Testing Your Agent', 'Anti-Patterns', and the Description/Body best-practice guides into a new reference file (e.g. references/agent-patterns.md) and link to it, leaving SKILL.md a lean overview.

Delete the redundant restatements: the skills-not-inherited point appears in both Anti-Patterns and Common Mistakes, and the plugin.json auto-discovery rule is repeated in Phase 7 and the Interaction Protocol.

Trim the 'Testing Your Agent' section to a short checklist; the four testing methods, symptoms table, iterative process, and tips over-explain testing fundamentals Claude already understands.

Dimension	Reasoning	Score
Conciseness	Noticeably verbose across ~1000 lines: 'Common Mistakes' (1-10) and 'Testing Your Agent' re-explain basic agent-design concepts Claude already knows, 'Interaction Protocol' restates the workflow, and the skills-not-inherited / plugin.json rules are repeated in multiple places.	2 / 5
Actionability	Fully executable guidance throughout — exact commands ('uvx skilllint@latest check', 'claude plugin validate', 'grep -q \"agents\"'), a complete copy-paste frontmatter template, concrete YAML tool patterns, and verbatim AskUserQuestion prompts.	5 / 5
Workflow Clarity	Eight clearly sequenced phases (Discovery → Post-Creation Validation) with explicit validation checkpoints, a Phase 6 checklist, and a fix-and-revalidate feedback loop ('If errors: fix and re-validate. Only when valid: pack').	5 / 5
Progressive Disclosure	References are clearly signaled one level deep in 'Quick Reference' and all three reference files exist, but large swaths of reference-grade material (Common Patterns, Anti-Patterns, Common Mistakes, Testing, Body/Description guides) are inlined in SKILL.md rather than split out.	3 / 5
	Total	15 / 20 Passed

Description

92%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong, third-person description that concretely states capabilities and gives explicit, natural trigger conditions. Minor improvement possible by trimming the 'high-quality' filler and adding synonym triggers.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — 'Create ... agents from scratch or by adapting existing agents as templates', 'Guides through requirements gathering, template selection, and agent file generation' — covering the domain comprehensively.	5 / 5
Completeness	Explicitly answers both what ('Create high-quality Claude Code agents ... Guides through requirements gathering, template selection, and agent file generation') and when ('Use when the user wants to create a new agent, modify agent configurations, build specialized subagents, or design agent architectures').	5 / 5
Trigger Term Quality	Strong natural triggers ('create a new agent', 'modify agent configurations', 'build specialized subagents', 'design agent architectures') that users would actually say, though no synonyms or variant phrasings are included.	4 / 5
Distinctiveness Conflict Risk	Occupies a clear niche (Claude Code agent authoring) with distinct triggers; the 'high-quality' buzzword aside, it is unlikely to fire for unrelated skills.	5 / 5
	Total	19 / 20 Passed

Validation

87%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (1007 lines); consider splitting into references/ and linking	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	14 / 16 Passed

Repository: Jamie-BitFlight/claude_skills
Path: plugins/plugin-creator/skills/agent-creator/SKILL.md
Commit: c337daf

Reviewed: 3 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.