claude-extensibility

Claude Code extensibility: agents, skills, output styles. Capabilities: create/update/delete agents and skills, YAML frontmatter, system prompts, tool/model selection, resumable agents, CLI-defined agents. Actions: create, edit, delete, optimize, test extensions. Keywords: agent, skill, output-style, SKILL.md, subagent, Task tool, progressive disclosure. Use when: creating agents/skills, editing extensions, configuring tool access, choosing models, testing activation.

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

75%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-organized, actionable reference that uses tables and code blocks effectively and correctly routes deep detail to two real reference files. Main weaknesses are minor content repetition in the Quick Reference and workflows with checkpoints that could be more explicit validation loops.

Suggestions

Remove or shrink the 'Quick Reference' section that duplicates model-selection and file-location tables already shown earlier, to recover tokens.

Add explicit validation/verification feedback loops to the Create/Optimize workflows (e.g. 'if activation test fails, revise the description and re-test') to strengthen workflow clarity for the destructive delete capability.

Move the built-in-subagents table and location/priority tables into the reference files to deepen progressive disclosure and shorten SKILL.md.

Dimension	Reasoning	Score
Conciseness	The body is dense and table/code-driven without explaining basics Claude already knows, but the 'Quick Reference' section restates model-selection and file-location content already covered above, which is minor padding that could be trimmed.	4 / 5
Actionability	Provides copy-paste-ready YAML templates, a concrete CLI-defined-agents JSON command, and management commands, though templates still contain placeholders like '[use case]' leaving minor gaps.	4 / 5
Workflow Clarity	Common Workflows give clear numbered sequences for Create Agent, Create Skill, and Optimize Extension, each ending with a test/verify checkpoint; validation is present but light, so minor gaps remain rather than full feedback loops.	4 / 5
Progressive Disclosure	Quick-start content is inline with two real, well-signaled one-level-deep references ('references/agent-development.md', 'references/skill-development.md'), but the body still inlines a fair amount of detail (built-in subagents, locations, CLI agents) that is well organized rather than maximally split.	4 / 5
	Total	16 / 20 Passed

Description

88%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, well-structured description that clearly states capabilities and explicit 'Use when' triggers in third person. Slightly held back by missing synonyms/file extensions and the inherent breadth of the 'agent'/'skill' trigger terms.

Dimension	Reasoning	Score
Specificity	Lists many concrete actions — 'create/update/delete agents and skills', 'edit, delete, optimize, test extensions', 'YAML frontmatter, system prompts, tool/model selection, resumable agents, CLI-defined agents' — giving comprehensive coverage rather than vague abstractions.	5 / 5
Completeness	Explicitly answers both what ('Claude Code extensibility: agents, skills, output styles' plus a Capabilities list) and when ('Use when: creating agents/skills, editing extensions, configuring tool access, choosing models, testing activation.').	5 / 5
Trigger Term Quality	Keywords 'agent, skill, output-style, SKILL.md, subagent, Task tool, progressive disclosure' give good natural coverage, but synonyms and file extensions (e.g. .claude/agents/) are missing, so it stops short of comprehensive.	4 / 5
Distinctiveness Conflict Risk	The Claude Code extensibility meta-niche is clear and distinct, but the trigger terms 'agent' and 'skill' are broadly used and carry minor overlap risk with other skills that mention those words.	4 / 5
	Total	18 / 20 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: samhvw8/dot-claude
Path: skills/claude-extensibility/SKILL.md
Commit: 28c7616

Reviewed: 23 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.