CtrlK
BlogDocsLog inGet started
Tessl Logo

claude-code-plugin-development

This skill should be used when the user asks to "create a plugin", "build a plugin", "write a plugin", or wants to bundle agents, hooks, commands, skills, or MCP servers into a distributable Claude Code plugin.

89

1.93x
Quality

86%

Does it follow best practices?

Impact

93%

1.93x

Average score across 3 eval scenarios

SecuritybySnyk

Failed to scan

The risk profile of this skill

SKILL.md
Quality
Evals
Security

Quality

Content

82%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, highly actionable reference skill with concrete schemas, commands, and troubleshooting. The main improvements would be removing inline duplication of content already covered by the reference files and adding an explicit validation feedback loop to the workflow.

Suggestions

Remove or shrink the inline 'CLI Quick Reference' and 'Plugin Components Summary' sections since they duplicate cli-commands.md and plugin-components.md; keep only a pointer to those references to improve progressive_disclosure and conciseness.

Add an explicit numbered workflow with a validate→fix→retry loop (e.g., build → install → `claude --debug` verify → on failure consult debugging.md and retry) to lift workflow_clarity to the top anchor.

Tighten the Permissions 'Problem/Cause/Solution' prose into a single compact pattern + solution so every remaining token earns its place.

DimensionReasoningScore

Conciseness

The body is dense and reference-style with minimal prose padding, using tables and code blocks and avoiding explanations of concepts Claude already knows; a few sections (Permissions Problem/Cause/Solution prose, the 'Important' callout) could be trimmed, fitting 'efficient; minor instances of over-explanation' rather than the fully lean top anchor.

4 / 5

Actionability

Provides copy-paste ready plugin.json and marketplace.json schemas, concrete directory structure, exact CLI commands, and specific JSON snippets for troubleshooting, covering the common cases — matching the 'fully executable; copy-paste ready' top anchor.

5 / 5

Workflow Clarity

A final Checklist sequences the work with a verification checkpoint ('Test with `claude --debug` to verify loading') and the Common Issues section supplies error-recovery guidance, fitting 'clear sequence with most checkpoints present; minor validation gaps'; not a 5 because there is no explicit validate→fix→retry feedback loop stated as a process.

4 / 5

Progressive Disclosure

The Quick Reference clearly signals four real one-level-deep reference files ('You MUST read these references'), but the body inlines content that overlaps those references (CLI Quick Reference duplicates cli-commands.md; Plugin Components Summary duplicates plugin-components.md), a minor organization gap that fits the 'good structure; references mostly clear; minor organization gaps' anchor rather than the fully split top anchor.

4 / 5

Total

17

/

20

Passed

Description

90%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 description that clearly states both capability and explicit trigger conditions with natural user phrasing. The only soft spot is specificity, since the described action is a single bundling operation rather than a list of distinct capabilities.

DimensionReasoningScore

Specificity

Names the domain (Claude Code plugins) and one concrete action — 'bundle agents, hooks, commands, skills, or MCP servers into a distributable Claude Code plugin' — but the action set is essentially a single bundling operation rather than multiple distinct actions, so it fits the 'names domain and 1-2 concrete actions' anchor rather than the 'lists several specific actions' anchor above.

3 / 5

Completeness

Explicitly answers both what ('bundle ... into a distributable Claude Code plugin') and when ('This skill should be used when the user asks to...') with concrete trigger phrases, matching the top anchor; uses third-person voice so no voice penalty applies.

5 / 5

Trigger Term Quality

Includes the exact natural phrases users would say — 'create a plugin', 'build a plugin', 'write a plugin' — covering synonyms comprehensively, so it matches the 'comprehensive coverage of natural terms including synonyms' anchor; not a 4 because no common natural trigger is missing for this domain.

5 / 5

Distinctiveness Conflict Risk

'Distributable Claude Code plugin' is a clear niche with distinct create/build/write triggers and minimal overlap with other skills, matching the 'clear niche with distinct triggers; minimal conflict risk' anchor.

5 / 5

Total

18

/

20

Passed

Validation

93%

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

Validation15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

allowed_tools_field

'allowed-tools' contains unusual tool name(s)

Warning

Total

15

/

16

Passed

Repository
dwmkerr/claude-toolkit
Reviewed

Table of Contents

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.