agent-development

This skill should be used when the user asks to "create an agent", "add an agent", "write a subagent", "agent frontmatter", "when to use description", "agent examples", "agent tools", "agent colors", "autonomous agent", or needs guidance on agent structure, system prompts, triggering conditions, or agent development best practices for Claude Code plugins.

Quality

70%

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-dev/skills/agent-development/SKILL.md

Quality

Content

63%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.

A well-structured, largely actionable reference for authoring Claude Code plugin agents, with concrete templates, field rules, and a validating implementation workflow. Its weaknesses are noticeable redundancy across the DO/DON'T, validation, and workflow sections, and a bundle that does not match its own references: three of six cited files are absent.

Suggestions

Create the missing examples/agent-creation-prompt.md and examples/complete-agent-examples.md files (or remove those references), and either add scripts/test-agent-trigger.sh or drop it from the 'Utility Scripts' section so all bundle pointers resolve.

Consolidate the redundant guidance: the final 'Best Practices' DO/DON'T lists, the 'Validation Rules' section, and the 'Implementation Workflow' repeat content already covered under 'Frontmatter Fields' and 'System Prompt Design' — merge them into one location to reduce token load.

Add a feedback loop to the workflow for the failure case, e.g. 'If scripts/validate-agent.sh fails, fix the reported field errors and re-run until validation passes before testing triggers.'

Dimension	Reasoning	Score
Conciseness	The body is dense with domain-specific guidance (frontmatter rules, color semantics, validation constraints), but the closing 'Best Practices' DO/DON'T lists, the 'Validation Rules' section, and the 'Implementation Workflow' largely restate points already made in earlier sections; it matches anchor 3 (mostly efficient, could be tightened) rather than 2.	3 / 5
Actionability	Concrete complete-format and minimal templates, exact field formats and length limits, a copy-paste-ready AI-assisted generation prompt, and script invocations ('Validate with scripts/validate-agent.sh') give mostly executable guidance; the placeholder-laden templates and missing example files leave minor gaps, so anchor 4 rather than 5.	4 / 5
Workflow Clarity	The 'Implementation Workflow' provides a clear 9-step sequence with explicit validation (step 7, validate-agent.sh) and testing (step 8) checkpoints, but there is no error-recovery guidance for what to do when validation or trigger tests fail — anchor 4 (most checkpoints, minor validation gaps) rather than 5's explicit feedback loops.	4 / 5
Progressive Disclosure	Structure and reference signaling are good ('Additional Resources' describes each file), and all three references/ files exist, but three of the six referenced bundle paths are dead — the examples/ directory does not exist (agent-creation-prompt.md, complete-agent-examples.md) and scripts/test-agent-trigger.sh is missing — breaking navigation for a third of the pointers, which pulls it below anchor 4's 'minor organization gaps'.	3 / 5
	Total	14 / 20 Passed

Description

77%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 trigger-focused description with excellent natural phrasing coverage and an explicit 'should be used when' clause, properly scoped to Claude Code plugins. Its main weakness is that the 'what' is expressed as vague 'guidance on...' topics rather than concrete actions, leaving it slightly below the top anchors.

Dimension	Reasoning	Score
Specificity	The description names the domain (agent development for Claude Code plugins) and several topic areas — 'agent structure, system prompts, triggering conditions, or agent development best practices' — but 'guidance on...' is not a list of concrete actions; it matches anchor 3 (domain plus 1-2 concrete actions, not comprehensive) better than anchor 4's several specific actions.	3 / 5
Completeness	Both what ('needs guidance on agent structure, system prompts, triggering conditions, or agent development best practices for Claude Code plugins') and when ('This skill should be used when the user asks to...') are present, but the what is folded into the when-clause rather than a distinct leading statement of capability; anchor 4 fits — both present, one could be more explicit.	4 / 5
Trigger Term Quality	Comprehensive coverage of natural phrasings users would actually say — 'create an agent', 'add an agent', 'write a subagent', 'agent frontmatter', 'when to use description', 'agent tools', 'agent colors', 'autonomous agent' — including multiple synonyms of the same intent, matching the top anchor.	5 / 5
Distinctiveness Conflict Risk	The trailing scope 'for Claude Code plugins' plus niche triggers like 'agent frontmatter' and 'agent colors' carve a clear niche, but broad terms like 'autonomous agent' carry minor overlap risk with general agent-related skills; anchor 4 (mostly distinct) fits better than 5.	4 / 5
	Total	16 / 20 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: anthropics/claude-plugins-official
Path: plugins/plugin-dev/skills/agent-development/SKILL.md
Commit: ad30d62

Reviewed: 6 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.