ProSkillsLibraryFreePassScan

Guide

SKILL.md examples

8 real skills from the free ProSkills catalog. Each shows its frontmatter with notes, what it does well, and the score the scanner gives it. The scores are computed from the shipped files when the site is built.

How to read an example

Every SKILL.md starts with frontmatter between two lines of three dashes. name must be lowercase with hyphens and match the folder. description is what the agent uses to decide when to load the skill. Below the frontmatter is the body: the steps and rules the agent follows.

01 · AI Models

character-bible

Scan 100/100Grade A

Lock a face, body and style across renders.

---
name: character-bible
description: "Use when the user creates, names, reuses or continues an AI character, model or influencer, or says \"keep the same face\", \"same girl every time\", \"my character looks different each render\". Writes a character bible with fixed face anchors, body, age range and a negative list, then repeats it word for word in every image or video prompt. Reads profile.md for model name, look, style, setting and camera."
---
The frontmatter
  • name: 15 characters, lowercase with hyphens, same as its folder.
  • description: 410 characters. It opens with "Use when", so the trigger comes first.
  • Together they cost about 107 tokens in every session, whether the skill loads or not.
What it does well
  • The body is about 1,189 tokens, loaded only when the skill triggers.
  • 1 supporting file (CHANGELOG.md) read on demand, not up front.
  • Clear sections: Before you start, Workflow, Rules and thresholds, Worked example, Checklist, Output.
  • Reads details from profile.md instead of hard-coding them.
  • The scanner reports no findings.
See character-bible, its first 40 lines and how to install it

02 · AI Models

wardrobe-sets

Scan 100/100Grade A

Outfit and scene ideas matched to your style.

---
name: wardrobe-sets
description: "Use when the user needs outfit ideas, wardrobe, styling, a lookbook, or location and scene ideas for an AI character or model, or asks \"what should she wear\", \"where should I shoot this\". Produces outfit and setting pairs matched to the style in profile.md, described by fabric, fit and colour instead of brands. Reads profile.md for model name, look, style, setting and camera."
---
The frontmatter
  • name: 13 characters, lowercase with hyphens, same as its folder.
  • description: 382 characters. It opens with "Use when", so the trigger comes first.
  • Together they cost about 99 tokens in every session, whether the skill loads or not.
What it does well
  • The body is about 945 tokens, loaded only when the skill triggers.
  • 1 supporting file (CHANGELOG.md) read on demand, not up front.
  • Clear sections: Before you start, Workflow, Rules and thresholds, Worked example, Checklist, Output.
  • Reads details from profile.md instead of hard-coding them.
  • The scanner reports no findings.
See wardrobe-sets, its first 40 lines and how to install it

03 · Frontend Taste

taste-core

Scan 100/100Grade A

Spacing, type scale, one accent.

---
name: taste-core
description: "Use when the user builds, restyles, reviews or says a UI \"looks off\", \"generic\" or \"needs polish\": pages, components, dashboards, landing pages. Sets a spacing grid, type scale and single accent before writing any CSS, then enforces them in the code. Reads profile.md for brand, accent colour, vibe and framework."
---
The frontmatter
  • name: 10 characters, lowercase with hyphens, same as its folder.
  • description: 319 characters. It opens with "Use when", so the trigger comes first.
  • Together they cost about 83 tokens in every session, whether the skill loads or not.
What it does well
  • The body is about 951 tokens, loaded only when the skill triggers.
  • 1 supporting file (CHANGELOG.md) read on demand, not up front.
  • Clear sections: Before you start, Workflow, Vibe table, Banned unless the user asks, Worked example, Checklist.
  • Reads details from profile.md instead of hard-coding them.
  • The scanner reports no findings.
See taste-core, its first 40 lines and how to install it

04 · Motion & Video

hyperframes-starter

Scan 100/100Grade A

Your first composition in one prompt.

---
name: hyperframes-starter
description: "Use when the user wants to make a video from HTML with HyperFrames, asks for \"a launch video\", \"a short promo\" or \"my first composition\", or needs a scene plan for one. Plans three to six scenes sized and timed to the chosen format and length, writes the composition, previews it, then renders. Reads profile.md for brand, format, length and colours."
---
The frontmatter
  • name: 19 characters, lowercase with hyphens, same as its folder.
  • description: 356 characters. It opens with "Use when", so the trigger comes first.
  • Together they cost about 94 tokens in every session, whether the skill loads or not.
What it does well
  • The body is about 1,055 tokens, loaded only when the skill triggers.
  • 1 supporting file (CHANGELOG.md) read on demand, not up front.
  • Clear sections: Before you start, Workflow, Rules and thresholds, Worked example, Checklist, Output.
  • Reads details from profile.md instead of hard-coding them.
  • The scanner reports no findings.
See hyperframes-starter, its first 40 lines and how to install it

05 · Humanizer

humanizer

Scan 100/100Grade A

Strips the most common AI tells.

---
name: humanizer
description: "Use when the user writes, edits or pastes prose and says \"humanize this\", \"make it sound less like AI\", \"too robotic\", or before you hand over any post, email, landing copy or doc you drafted. Strips the most common AI tells: banned words, em-dash asides, not-X-but-Y framing, signposts and summary endings, in one pass without changing facts. Reads profile.md for brand, voice and never-say words."
---
The frontmatter
  • name: 9 characters, lowercase with hyphens, same as its folder.
  • description: 404 characters. It opens with "Use when", so the trigger comes first.
  • Together they cost about 104 tokens in every session, whether the skill loads or not.
What it does well
  • The body is about 1,024 tokens, loaded only when the skill triggers.
  • 1 supporting file (CHANGELOG.md) read on demand, not up front.
  • Clear sections: Before you start, Workflow, Structural tells to cut, Word tells to replace, Voice table, Do not over-correct.
  • Reads details from profile.md instead of hard-coding them.
  • The scanner reports no findings.
See humanizer, its first 40 lines and how to install it

06 · Viral Hooks

hook-lab

Scan 100/100Grade A

Five hooks per post, each a different pattern.

---
name: hook-lab
description: "Use when the user writes or rewrites the first line of a post, tweet, LinkedIn update, caption or video opener, or says \"I need a hook\", \"first line\", \"this post is not landing\". Writes five hooks per post, each a different pattern, each under twelve words and a specific claim. Reads profile.md for brand, audience and platforms."
---
The frontmatter
  • name: 8 characters, lowercase with hyphens, same as its folder.
  • description: 336 characters. It opens with "Use when", so the trigger comes first.
  • Together they cost about 87 tokens in every session, whether the skill loads or not.
What it does well
  • The body is about 971 tokens, loaded only when the skill triggers.
  • 1 supporting file (CHANGELOG.md) read on demand, not up front.
  • Clear sections: Before you start, Workflow, Patterns, Rules and thresholds, Worked example, Checklist.
  • Reads details from profile.md instead of hard-coding them.
  • The scanner reports no findings.
See hook-lab, its first 40 lines and how to install it

07 · Grill Me

grill-me

Scan 100/100Grade A

A relentless review of your plan.

---
name: grill-me
description: "Use when the user shares a plan, spec, feature or architecture, asks to build something non-trivial, or says \"grill me\", \"poke holes\", \"what am I missing\". Interrogates the plan one question at a time, starting with data, failure and scale, and writes no code until the questions are resolved. Reads profile.md for project, stack and risk appetite."
---
The frontmatter
  • name: 8 characters, lowercase with hyphens, same as its folder.
  • description: 354 characters. It opens with "Use when", so the trigger comes first.
  • Together they cost about 91 tokens in every session, whether the skill loads or not.
What it does well
  • The body is about 1,075 tokens, loaded only when the skill triggers.
  • 1 supporting file (CHANGELOG.md) read on demand, not up front.
  • Clear sections: Before you start, Depth, Workflow, What each branch must probe, Anti-patterns, Worked example.
  • Reads details from profile.md instead of hard-coding them.
  • The scanner reports no findings.
See grill-me, its first 40 lines and how to install it

08 · Code Review

pr-describer

Scan 100/100Grade A

Title and description from your diff. No filler.

---
name: pr-describer
description: "Use when the user says 'write the PR description', 'describe this PR', 'PR title', 'summarise my changes', 'what should the commit message be', or is about to open a pull request. Reads the diff and the commits, then writes a short, honest title and description: what changed, why, how to test it, risk, and a screenshots placeholder for UI changes. No filler. Reads profile.md for project, stack, strictness and conventions."
---
The frontmatter
  • name: 12 characters, lowercase with hyphens, same as its folder.
  • description: 425 characters. It opens with "Use when", so the trigger comes first.
  • Together they cost about 110 tokens in every session, whether the skill loads or not.
What it does well
  • The body is about 1,207 tokens, loaded only when the skill triggers.
  • 1 supporting file (CHANGELOG.md) read on demand, not up front.
  • Clear sections: Before you start, Workflow, Template, What, Why, How.
  • Reads details from profile.md instead of hard-coding them.
  • The scanner reports no findings.
See pr-describer, its first 40 lines and how to install it

Questions

What does a SKILL.md file contain?
Two parts. YAML frontmatter at the top with a name and a description, then a Markdown body with the instructions. The agent reads only the name and description until the task matches, then loads the body.
How long should a SKILL.md be?
Keep the body under 500 lines and about 5,000 tokens. Put long tables, catalogues and examples in separate files that the body points to, so they load only when needed.
Where does the description matter most?
It is the only text the agent sees before it decides to load the skill. Say what the skill does and when to use it, and name the words a user would actually type.
Can I copy these examples?
The skills on this page are free. Install one from its page, open the folder, and edit it. Each skill reads a profile.md for details, so you can change those without touching the instructions.

Keep reading

Scan your own skill

Paste a SKILL.md or drop a folder. It is scored in your browser and nothing is uploaded.

Open the Skill Scanner

Want a skill that already passes? Start with a free one.

Character BibleFreeScan 100/100Lock a face, body and style across renders. From the AI Models collection.