skill-creator

Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch for Claude Code or Cursor, update or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.

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

62%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 body is well-structured and highly actionable with a clear, validated workflow, but it is padded with conversational filler and contains references to bundle files (agents/*.md, eval-viewer/generate_review.py) that are not present, weakening navigation.

Suggestions

Remove conversational padding (e.g. 'Cool? Cool.', the plumber anecdote, 'billions a year in economic value') and explanations of concepts Claude already knows to tighten conciseness.

Create the missing referenced files (agents/grader.md, agents/comparator.md, agents/analyzer.md) and the eval-viewer/generate_review.py script, or update the body to point at the scripts/ files that actually exist.

Verify every path mentioned in the body resolves to a real bundle file so progressive-disclosure navigation is not broken.

Dimension	Reasoning	Score
Conciseness	Noticeably verbose with padded conversational asides ('Cool? Cool.', the plumber anecdote, 'we are trying to create billions a year in economic value here!') and explanation of concepts Claude already knows, beyond the 'mostly efficient' midpoint.	2 / 5
Actionability	Provides concrete executable commands (e.g. 'python -m scripts.aggregate_benchmark', 'python -m scripts.run_loop ...'), JSON templates, and exact field names; only minor gaps keep it from copy-paste-perfect.	4 / 5
Workflow Clarity	The eval process is clearly sequenced across Steps 1-5 with an explicit validation checkpoint ('Do NOT generate the viewer or benchmark until grading.json exists for every run') and a feedback-driven iteration loop.	5 / 5
Progressive Disclosure	Body is under 500 lines with one-level-deep, clearly signaled references, but several referenced paths (agents/grader.md, agents/comparator.md, agents/analyzer.md, eval-viewer/generate_review.py) do not exist in the bundle, so navigation is partly broken.	3 / 5
	Total	14 / 20 Passed

Description

100%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 exemplary: it states concrete capabilities, provides a thorough 'Use when' trigger clause with natural phrasings and tool names, and occupies a distinct niche with low conflict risk. No gaps relative to the rubric anchors.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions ('Create new skills, modify and improve existing skills, and measure skill performance'; 'run evals', 'benchmark skill performance with variance analysis', 'optimize a skill's description for better triggering accuracy') with comprehensive coverage.	5 / 5
Completeness	Explicitly answers both 'what' (create/modify/measure skills) and 'when' ('Use when users want to create a skill from scratch ... update or optimize ... run evals ... benchmark ... or optimize a skill's description').	5 / 5
Trigger Term Quality	Comprehensive natural terms including synonyms and tool names ('create a skill from scratch', 'Claude Code or Cursor', 'run evals', 'benchmark', 'variance analysis', 'optimize ... triggering accuracy').	5 / 5
Distinctiveness Conflict Risk	Clear niche of skill creation and optimization for Claude Code/Cursor with distinct triggers and minimal conflict risk with other skills.	5 / 5
	Total	20 / 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: cognitedata/builder-skills
Path: skills/skill-creator/SKILL.md
Commit: c87160a

Reviewed: 5 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.