create-agent

Comprehensive guide for creating Claude Code agents with proper structure, triggering conditions, system prompts, and validation - combines official Anthropic best practices with proven patterns

Quality

52%

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/customaize-agent/skills/create-agent/SKILL.md

Quality

Content

46%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 content is detailed and actionable with strong templates and worked examples, but it is notably verbose with redundant sections, inlines material that should be split into reference files, and references a validation script that is absent from the bundle.

Suggestions

Move the two full "Production Examples" and the frontmatter field reference into separate files under references/ and link to them from SKILL.md, turning the main file into a concise overview.

Remove or condense the "What Are Agents?" section and the agent-vs-command table, which explain concepts Claude already knows.

Either provide the referenced `scripts/validate-agent.sh` or replace the reference with inline validation checks so the stated workflow is actually runnable.

Dimension	Reasoning	Score
Conciseness	The ~660-line body is noticeably verbose: it explains concepts Claude already knows (the "What Are Agents?" section, agent-vs-command table) and repeats material (description/example rules, validation tables, and DO/DON'T lists that duplicate the checklist).	2 / 5
Actionability	It provides largely executable guidance — exact frontmatter format, a concrete system-prompt template, two full worked agent examples, and bash commands like `mkdir -p ${CLAUDE_PLUGIN_ROOT}/agents` — with only minor gaps such as the placeholder-laden templates.	4 / 5
Workflow Clarity	The six-step "Agent Creation Process" is sequenced and includes an explicit validation step with a checklist, but the referenced `scripts/validate-agent.sh` does not exist in the bundle and the validate→fix→retry feedback loop is only weakly implied.	3 / 5
Progressive Disclosure	Everything is inlined into a single monolithic file with no bundle files; the two full production examples, field reference, and templates clearly belong in separate reference files, and the one referenced path (validate-agent.sh) is broken.	2 / 5
	Total	11 / 20 Passed

Description

58%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 specific and distinct, clearly communicating the skill's domain and component topics, but it omits any "when to use" trigger guidance and relies on generic keywords rather than natural user phrasings.

Suggestions

Add a "Use when..." clause stating concrete trigger scenarios, e.g. "Use when the user asks to create or refine a Claude Code agent/subagent."

Include natural trigger terms users actually say, such as "create an agent", "set up a subagent", or "write an agent system prompt".

Trim the marketing phrase "combines official Anthropic best practices with proven patterns" which adds fluff without improving triggerability.

Dimension	Reasoning	Score
Specificity	The description names the domain ("creating Claude Code agents") and lists several concrete facets ("structure, triggering conditions, system prompts, and validation"), giving it strong but not exhaustive coverage of specific actions.	4 / 5
Completeness	It clearly answers "what" the skill does but provides no "Use when..." clause or equivalent trigger guidance, which per the judging guidelines caps completeness at 3.	3 / 5
Trigger Term Quality	It includes relevant keywords ("creating", "agents", "validation") but lacks the natural trigger phrases a user would actually say (e.g. "create an agent", "set up a subagent").	3 / 5
Distinctiveness Conflict Risk	The phrase "creating Claude Code agents" carves out a clear niche with distinct triggers, with only minor overlap risk against general plugin-dev skills.	4 / 5
	Total	14 / 20 Passed

Validation

75%

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 — 12 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (670 lines); consider splitting into references/ and linking	Warning
allowed_tools_field	'allowed-tools' contains unusual tool name(s)	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
referenced_paths_exist	Referenced path issues: 2 missing	Warning

	Total	12 / 16 Passed

Repository: NeoLabHQ/context-engineering-kit
Path: plugins/customaize-agent/skills/create-agent/SKILL.md
Commit: 23e2428

Reviewed: 18 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.