Guides writing a new agent skill: when one is warranted, the required folder layout and frontmatter, section order and size limits, with a bundled style guide.

Apache-2.0Auto-check passedAgent Workflows

Install Skill Creator

skills CLI
$ npx skills add Gentleman-Programming/gentle-ai --skill skill-creator -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install Gentleman-Programming/gentle-ai skill-creator --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/Gentleman-Programming/gentle-ai.git skills-src && mkdir -p .claude/skills && cp -r skills-src/internal/assets/skills/skill-creator .claude/skills/skill-creator && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
skill-creator
GitHub stars
7.6k
Token cost
~998 tokens
SKILL.md length
394 words
Files
2 (incl. references)
Skills in repo
15
Repo updated
First seen
Licence
Apache-2.0

At a glance

Guides writing a new agent skill: when one is warranted, the required folder layout and frontmatter, section order and size limits, with a bundled style guide.

  • Works in 7 steps: Check whether docs/skill-style-guide.md… → If the repo guide is unavailable, read… → Confirm the skill does not already exist… → …
  • Writing a new skill for a workflow the agent keeps getting wrong
  • SKILL.md covers Activation Contract, Hard Rules, Decision Gates and Execution Steps, plus 3 more sections
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

This skill covers when and how to create a new skill. A skill is worth writing when a pattern repeats and the agent needs guidance, project conventions differ from generic practice, a workflow needs step-by-step instructions or a decision tree helps pick an approach. Trivial or one-off patterns are better left to normal documentation. It treats a skill as a runtime instruction contract for an LLM, not as human documentation.

The steps check for the repo's docs/skill-style-guide.md, fall back to the bundled references/skill-style-guide.md and then to inline rules, confirm the skill does not already exist, and create a SKILL.md under the skills folder with name and description frontmatter that opens with a trigger. Sections follow a fixed order: Activation Contract, Hard Rules, Decision Gates, Execution Steps, Output Contract and References. Templates go in assets, detail in references, there is no Keywords section, and the body stays short, around 180 to 450 tokens with a hard maximum of 1000. Project skills are registered in AGENTS.md.

When your agent uses it

  • Writing a new skill for a workflow the agent keeps getting wrong
  • Deciding whether a pattern deserves a skill or just documentation
  • Checking a skill's frontmatter and section order against the style guide

Example prompts

  • “Create a skill that captures how we name database migrations in this repo.”
  • “Is our release checklist worth turning into a skill, or should it stay in the docs?”
  • “Review skills/api-errors/SKILL.md against the style guide and fix the frontmatter.”

Workflow steps

7 steps, taken from the first numbered list in SKILL.md.

  1. Check whether docs/skill-style-guide.md exists; if it does, apply it before the bundled local copy or inline fallback rules.
  2. If the repo guide is unavailable, read references/skill-style-guide.md and apply it before the inline fallback rules.
  3. Confirm the skill does not already exist and the pattern is reusable.
  4. Create or update skills/{skill-name}/SKILL.md using this required structure
  5. Use this frontmatter shape
  6. Write sections in this order: Activation Contract, Hard Rules, Decision Gates, Execution Steps, Output Contract, References.
  7. Register the skill in AGENTS.md when it is a project skill.

What it can do on your machine

Read from SKILL.md and the folder at commit 81d2225. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md (its code samples are yaml and markdown).

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Skill Creator loads about 998 tokens when it runs, and up to ~1.8k if it reads all its reference files. Until then it costs about 33 tokens; SKILL.md has 394 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~33
When it runs · the whole SKILL.md, loaded when a task matches
~998
With references · SKILL.md plus every file in references/, read only if the agent opens them
~1.8k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from Gentleman-Programming/gentle-ai at commit 81d2225, republished under its Apache-2.0 licence (© Gentleman-Programming). 394 words, ~998 tokens.

Download SKILL.mdSave it as .claude/skills/skill-creator/SKILL.md (or your agent's skills folder). This skill also uses 1 other file; get the full folder from GitHub.
name
skill-creator
description
Trigger: new skills, agent instructions, documenting AI usage patterns. Create LLM-first skills with valid frontmatter.
license
Apache-2.0
metadata.author
gentleman-programming
metadata.version
1.0

Activation Contract

Create a skill when:

  • A pattern is used repeatedly and AI needs guidance
  • Project-specific conventions differ from generic best practices
  • Complex workflows need step-by-step instructions
  • Decision trees help AI choose the right approach

Do not create a skill when the pattern is trivial, one-off, or better served by normal documentation.

Hard Rules

  • When working in this repo, first follow docs/skill-style-guide.md as the normative source before creating or updating skills.
  • For installed global skills, use references/skill-style-guide.md as the bundled local copy of that guide when docs/skill-style-guide.md is unavailable.
  • If neither guide is available, use the compact inline rules below.
  • A skill is a runtime instruction contract for an LLM, not human documentation.
  • Do not add a Keywords section; preserve essential trigger words in description.
  • References must point to local files.
  • Keep the skill body concise: target 180–450 tokens, recommended max 700, hard max 1000.

Decision Gates

NeedAction
Code templates, schemas, fixtures, generated examplesPut them in assets/
Conceptual detail, edge cases, existing docsPut local links in references/
Long explanation in SKILL.mdMove it to a supporting file
Multiple meaningful pathsAdd a compact decision table

Execution Steps

  1. Check whether docs/skill-style-guide.md exists; if it does, apply it before the bundled local copy or inline fallback rules.
  2. If the repo guide is unavailable, read references/skill-style-guide.md and apply it before the inline fallback rules.
  3. Confirm the skill does not already exist and the pattern is reusable.
  4. Create or update skills/{skill-name}/SKILL.md using this required structure:
skills/{skill-name}/
├── SKILL.md              # Required - main skill file
├── assets/               # Optional - templates, schemas, examples
│   ├── template.py
│   └── schema.json
└── references/           # Optional - links to local docs
    └── docs.md           # Points to docs/developer-guide/*.mdx
  1. Use this frontmatter shape:
markdown
---
name: {skill-name}
description: "Trigger: {essential trigger words users or agents will say}. {What this skill does}."
license: Apache-2.0
metadata:
  author: "{your-github-username}"
  version: "1.0"
---
  1. Write sections in this order: Activation Contract, Hard Rules, Decision Gates, Execution Steps, Output Contract, References.
  2. Register the skill in AGENTS.md when it is a project skill.
Show full SKILL.md (112 more words)Show less

Inline Fallback Rules

  • description MUST be one physical line, quoted, YAML-safe, and include essential trigger words first.
  • description SHOULD be <=160 chars and MUST be <=250 chars.
  • Frontmatter MUST include name, description, license, metadata.author, and metadata.version.
  • Use imperative instructions, not tutorials or background prose.
  • Put supporting material in assets/ or references/, not the main skill body.

Good:

yaml
description: "Trigger: Jira task, ticket, issue, task creation. Create Jira tasks in the team format."

Bad:

yaml
description: >
  Create Jira tasks in the team format.
  Trigger: Jira task, ticket, issue, or task creation.
Keywords: jira, task

Output Contract

Return:

  • Files created or modified.
  • Whether the repo style guide or inline fallback rules were used.
  • Any AGENTS.md registration change.
  • Any supporting files added under assets/ or references/.

References

  • docs/skill-style-guide.md — normative LLM-first skill style guide for this repo.
  • references/skill-style-guide.md — bundled local copy for installed global skills when the repo doc is unavailable.

© Gentleman-Programming, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 1 other file (references) in internal/assets/skills/skill-creator of Gentleman-Programming/gentle-ai.

  • SKILL.md
  • references/skill-style-guide.md

Open the folder on GitHubat commit 81d2225

Compare with similar skills

Skill Creator next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Skill Creator compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Skill Creator this skillGentleman-Programming/gentle-ai7.6k—~998Automated safety check: PassApache-2.0
Harness Agent Team Designerrevfactory/harness9.1k—~4.5kAutomated safety check: PassApache-2.0
Harness Evolution Feedback Looprevfactory/harness9.1k—~855Automated safety check: PassApache-2.0
Skill Creatorccusage/ccusage19k—~1.1kAutomated safety check: PassCustom licence
ShellLM Skill Authorlaude-institute/headlong1.2k—~1.3kAutomated safety check: PassApache-2.0
Fix Prompt Evaluation Diagnosticsmicrosoft/vscode-chat-customizations-evaluation141—~506Automated safety check: PassCustom licence

Similar skills

  • Harness Agent Team Designer

    revfactory/harness

    Designs a project-specific agent harness: defines specialist agents, writes the skills they follow, picks an execution mode and model for each, and keeps the setup maintained.

    9.1k GitHub stars~4.5k tokensUpdated 9 days ago
    Agent WorkflowsAuto-check passed
  • Collects feedback on how an agent harness performed, generalizes it, and updates the harness agents, skills and orchestrator along with a change-history table.

    9.1k GitHub stars~855 tokensUpdated 9 days ago
    Agent WorkflowsAuto-check passed
  • Skill Creator

    ccusage/ccusage

    Guides repo-local skill creation and updates. An agent skill from ccusage/ccusage.

    19k GitHub stars~1.1k tokensUpdated today
    Agent WorkflowsAuto-check passed
  • ShellLM Skill Author

    laude-institute/headlong

    Scaffolds new ShellLM skills with the right frontmatter, directory layout and agent-facing writing style, so an agent can extend its own capabilities.

    1.2k GitHub stars~1.3k tokensUpdated today
    Agent WorkflowsAuto-check passed
  • Fix Prompt Evaluation Diagnostics

    microsoft/vscode-chat-customizations-evaluation

    Official

    Applies the fixes suggested by the Chat Customizations Evaluations extension to a prompt, agent, skill or instruction file, changing only what each diagnostic requires.

    141 GitHub stars~506 tokensUpdated 4 days ago
    Agent WorkflowsAuto-check passed
  • Skill Creator

    Gentleman-Programming/Gentleman-Skills

    Creates new AI agent skills following the Agent Skills spec.

    657 GitHub stars~1.1k tokensUpdated 6 mo ago
    Agent WorkflowsAuto-check: notes

More from Gentleman-Programming/gentle-ai

All 15 skills in this repo
  • GitHub Issue Creation

    Gentleman-Programming/gentle-ai

    Drafts, creates, comments on and approves GitHub issues under strict rules: YAML Issue Forms, a duplicate search first, and guarded protected labels.

    7.6k GitHub stars~3.3k tokensUpdated today
    Auto-check passed
  • Gentle AI Pull Requests

    Gentleman-Programming/gentle-ai

    Prepares pull requests for the Gentle AI project under an issue-first rule: a linked approved issue, one type label, a valid branch name and confirmed required CI.

    7.6k GitHub stars~2.3k tokensUpdated today
    Auto-check passed
  • Work-Unit Commits

    Gentleman-Programming/gentle-ai

    Plans commits and PRs as reviewable work units, keeping tests and docs with the code they cover and splitting large changes into chained PRs.

    7.6k GitHub stars~1.1k tokensUpdated today
    Auto-check passed
  • Go Testing

    Gentleman-Programming/gentle-ai

    Trigger: Go tests, go test coverage, Bubbletea teatest, golden files.

    7.6k GitHub stars~550 tokensUpdated today
    Auto-check passed
  • Hermes Ephemeral Delegation

    Gentleman-Programming/gentle-ai

    Trigger: a mapping need, parallel units, context backstop, high-risk verify, fresh review, or multi-step debug.

    7.6k GitHub stars~1.1k tokensUpdated today
    Auto-check passed
  • Judgment Day

    Gentleman-Programming/gentle-ai

    Trigger: judgment day, dual review, adversarial review, juzgar.

    7.6k GitHub stars~927 tokensUpdated today
    Auto-check passed

Categories

Questions about Skill Creator

What does Skill Creator do?

Guides writing a new agent skill: when one is warranted, the required folder layout and frontmatter, section order and size limits, with a bundled style guide. This skill covers when and how to create a new skill. A skill is worth writing when a pattern repeats and the agent needs guidance, project conventions differ from generic practice, a workflow needs step-by-step instructions or a decision tree helps pick an approach.

When should I use Skill Creator?

Skill Creator fits situations like: writing a new skill for a workflow the agent keeps getting wrong; deciding whether a pattern deserves a skill or just documentation; checking a skill's frontmatter and section order against the style guide.

How do I install Skill Creator in Claude Code?

Run `npx skills add Gentleman-Programming/gentle-ai --skill skill-creator -a claude-code`. Or copy the skill folder (internal/assets/skills/skill-creator in Gentleman-Programming/gentle-ai) into .claude/skills/skill-creator in your project. Claude Code loads it when a task matches its description.

How do I install Skill Creator in Codex?

Run `npx skills add Gentleman-Programming/gentle-ai --skill skill-creator -a codex`. Or copy the skill folder (internal/assets/skills/skill-creator in Gentleman-Programming/gentle-ai) into .agents/skills/skill-creator in your project. Codex loads it when a task matches its description.

Can I use Skill Creator in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add Gentleman-Programming/gentle-ai --skill skill-creator -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/skill-creator, .gemini/skills/skill-creator, .github/skills/skill-creator and .opencode/skills/skill-creator in your project.

What does Skill Creator need to run?

SKILL.md names no scripts, command-line tools or credentials: Skill Creator is instructions for the agent only.

Does Skill Creator access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Skill Creator safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Skill Creator use?

Skill Creator is published under the Apache-2.0 licence (declared in SKILL.md). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Skill Creator use?

About 998 tokens (SKILL.md is roughly 4k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 843 tokens, read only when the agent opens those files.

What are the alternatives to Skill Creator?

Skills that share tags, products or a category with Skill Creator: Harness Agent Team Designer (revfactory/harness, 9.1k stars), Harness Evolution Feedback Loop (revfactory/harness, 9.1k stars), Skill Creator (ccusage/ccusage, 19k stars) and ShellLM Skill Author (laude-institute/headlong, 1.2k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Skill Creator?

Gentleman-Programming (a GitHub organization) maintains it in Gentleman-Programming/gentle-ai, which has 7,583 GitHub stars. The repository holds 15 skills in this directory. The repository was last updated on October 7, 2026.

Source: Gentleman-Programming/gentle-ai on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.