Guide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends Claude's capabilities with specialized knowledge, workflows, or tool integrations.
Install with Tessl CLI
npx tessl i github:vercel-labs/agent-browser --skill skill-creatorOverall
score
81%
Does it follow best practices?
If you maintain this skill, you can automatically optimize it using the tessl CLI to improve its score:
npx tessl skill review --optimize ./path/to/skillEvaluation — 79%
↑ 1.23xAgent success when using this skill
Validation for skill structure
New skill initialization and creation workflow
Init script usage
0%
100%
SKILL.md exists
0%
100%
Correct directory structure
0%
100%
No extraneous docs
100%
100%
Frontmatter has name
0%
100%
Frontmatter has description
0%
100%
No extra frontmatter fields
0%
100%
Name is hyphen-case
100%
100%
Scripts directory present
0%
100%
References directory present
0%
100%
Example files cleaned up
100%
100%
Log documents steps
100%
100%
Effective SKILL.md writing and frontmatter
Description includes triggers
0%
100%
No 'when to use' in body
100%
100%
Description under 1024 chars
0%
100%
No angle brackets in description
0%
100%
Only name and description in frontmatter
0%
100%
Imperative voice in body
60%
100%
Concise skill body
100%
100%
Examples over explanations
20%
100%
Style guide in references
0%
100%
Template in assets
0%
100%
References linked from SKILL.md
100%
100%
Progressive disclosure and resource organization
Separate provider reference files
0%
100%
SKILL.md under 500 lines
100%
100%
References linked from SKILL.md
100%
100%
No content duplication
100%
100%
Flat reference structure
0%
100%
TOC in long reference files
0%
0%
Cross-provider policies separate
100%
100%
Terraform scaffold script
100%
100%
Selection guidance in SKILL.md
100%
100%
No extraneous files
100%
100%
Degrees of freedom and workflow design
Low freedom for column drops
100%
0%
High freedom for new tables
100%
0%
Medium freedom for indexes
100%
0%
Sequential workflow for dangerous ops
100%
0%
Script for fragile operation
100%
0%
Conditional workflow routing
0%
0%
Concise body
100%
100%
Imperative form
100%
0%
Design rationale explains freedom levels
100%
0%
No extraneous docs in skill
100%
100%
Tested scripts
0%
0%
Iterating and packaging existing skill
README.md removed
100%
100%
CHANGELOG.md removed
100%
100%
Example script removed
100%
100%
Placeholder reference removed
100%
100%
Placeholder asset removed
100%
100%
Extra frontmatter fields removed
100%
100%
Description improved
100%
100%
When to use section removed from body
100%
100%
Content split into references
100%
100%
SKILL.md under 500 lines
100%
100%
Packaging mentioned
0%
0%
References linked from SKILL.md
100%
100%
Table of Contents
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.