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

79%

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

Fix and improve this skill with Tessl

tessl review fix ./plugins/plugin-creator/skills/agent-creator/SKILL.md

Quality

Content

66%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.

The skill is highly actionable with a well-sequenced, validated workflow and clear reference-file navigation. Its main weakness is verbosity: large sections restate general knowledge (testing methodology, common mistakes) that inflates the token budget without adding skill-specific value.

Suggestions

Move the 'Testing Your Agent' section into a reference file (e.g. references/agent-testing.md) and keep only a brief pointer in SKILL.md, since the bulk restates general testing methodology Claude already knows.

Condense the 10 'Common Mistakes' entries into a short checklist or relocate them to a reference file; retain only the skill-specific pitfalls inline.

Trim the Description Writing Guide and Agent Body Best Practices sections to the essential template plus one example each, removing redundant explanations of what action verbs and XML tags are.

Dimension	Reasoning	Score
Conciseness	The body is noticeably verbose: the 'Common Mistakes' (10 detailed entries) and 'Testing Your Agent' sections (~140 lines covering generic testing methodology, edge-case testing, iterative process, tips) explain concepts Claude already knows and could be trimmed or moved to a reference file.	2 / 5
Actionability	Concrete, executable guidance is abundant — bash commands ('ls -la .claude/agents/', 'uvx skilllint@latest check ...', 'claude plugin validate ...'), full agent-file templates, and YAML tool-pattern examples — with only minor template placeholders ({...}) leaving small gaps.	4 / 5
Workflow Clarity	An 8-phase workflow is clearly sequenced with explicit validation checkpoints (Phase 6 checklist, Phase 8 post-creation validation), executable verification ('VERIFY file created successfully', 'RUN validation'), and a fix-and-revalidate feedback loop for file/plugin creation.	5 / 5
Progressive Disclosure	Quick Reference clearly signals one-level-deep reference files (agent-schema.md, agent-templates.md, agent-examples.md) which all exist, but sizable inlined sections (Common Mistakes, Testing) that could live in separate files keep it just short of ideal.	4 / 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.

The description is third-person, specific, and fully answers both 'what' and 'when' with concrete, natural trigger phrases. It is a strong, low-conflict description; only minor synonym coverage keeps trigger_term_quality from a perfect score.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — 'Create high-quality Claude Code agents from scratch or by adapting existing agents as templates', 'modify agent configurations, build specialized subagents, design agent architectures', and 'Guides through requirements gathering, template selection, and agent file generation' — giving comprehensive coverage.	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') with concrete triggers.	5 / 5
Trigger Term Quality	Natural trigger phrases like 'create a new agent', 'modify agent configurations', 'build specialized subagents', and 'design agent architectures' are present and user-sayable, though a few synonymous phrasings are missing.	4 / 5
Distinctiveness Conflict Risk	It carves a clear niche (creating Claude Code agents) with distinct, specific triggers and minimal overlap risk with other 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 (1003 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: 4e1e73b

Reviewed: ~14 hours 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.