How do I install the Mass Skill Authoring skill?

Run npx skillmds@latest add maiconsouza89/mass-skill-authoring in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Mass Skill Authoring safe to use?

This skill has not completed SkillMD's automated safety review yet. Capability flags: docs only. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Mass Skill Authoring?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Mass Skill Authoring free to use?

Yes. Installing skills from SkillMD is free. This skill is licensed under CC-BY-4.

Who published Mass Skill Authoring?

maiconsouza89 (@maiconsouza89) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Mass Skill Authoring by maiconsouza89

Name: Mass Skill Authoring
Rating: 4.2 (5 reviews)
Author: maiconsouza89

Mass Skill Authoring

Writes or reviews a Mass Solutions skill so it passes the catalog validator on the first run. Use when "create a new skill", "review this SKILL.md" or "why does the validator reject my skill". Do NOT use for writing commit messages or pull requests (use mass-commit-message or mass-pr-description).

maiconsouza89 Updated Sep 22, 2026

File contents

Skill authoring

A skill is a folder under skills/ with one SKILL.md and, optionally, references/, scripts/, assets/ and evals/. The validator enforces the contract in references/skill-contract.md; read it once, then follow the steps below.

Steps

Scaffold: pnpm new-skill mass-<slug>. The name equals the folder, is kebab-case, starts with mass- and has at most 64 characters.
Write the description with the formula and nothing else: [What it does]. Use when "a", "b" or "c". Do NOT use for X (use mass-y). The "Use when" quotes are the phrases a user would actually type. The "Do NOT use for" clause names the neighbouring skill so the two never overlap.
Keep SKILL.md under 80 lines for example skills and under 500 lines for any skill. Move long material to references/ and link it with a relative path.
Put executable code in scripts/ with a #! first line and the executable bit set (chmod +x). Put templates in assets/.
Fill metadata: author, version (semver), category (an id from skills/_categories.json), tags (comma-separated string) and reviewed (today, as YYYY-MM-DD). Every value is a string.
Optionally add evals/triggers.json with should and shouldNot prompt lists to record what must and must not trigger the skill.
Run pnpm check. Fix every reported <rule> <path>:<line> before opening a PR.

Review checklist

The description answers "what", "when" and "not when" in one paragraph.
No secrets, no binaries, no downloads piped into a shell, no instructions that ask the agent to hide information from the user.
Every relative link in SKILL.md resolves to a file inside the skill.
metadata.reviewed was updated in this change.

---name: mass-skill-authoringdescription: Writes or reviews a Mass Solutions skill so it passes the catalog validator on the first run. Use when "create a new skill", "review this SKILL.md" or "why does the validator reject my skill". Do NOT use for writing commit messages or pull requests (use mass-commit-message or mass-pr-description).license: CC-BY-4.0---# Skill authoringA skill is a folder under `skills/` with one `SKILL.md` and, optionally, `references/`,`scripts/`, `assets/` and `evals/`. The validator enforces the contract in[references/skill-contract.md](references/skill-contract.md); read it once, then follow thesteps below.## Steps1. Scaffold: `pnpm new-skill mass-<slug>`. The name equals the folder, is kebab-case, starts with `mass-` and has at most 64 characters.2. Write the `description` with the formula and nothing else: `[What it does]. Use when "a", "b" or "c". Do NOT use for X (use mass-y).` The "Use when" quotes are the phrases a user would actually type. The "Do NOT use for" clause names the neighbouring skill so the two never overlap.3. Keep `SKILL.md` under 80 lines for example skills and under 500 lines for any skill. Move long material to `references/` and link it with a relative path.4. Put executable code in `scripts/` with a `#!` first line and the executable bit set (`chmod +x`). Put templates in `assets/`.5. Fill `metadata`: `author`, `version` (semver), `category` (an id from `skills/_categories.json`), `tags` (comma-separated string) and `reviewed` (today, as `YYYY-MM-DD`). Every value is a string.6. Optionally add `evals/triggers.json` with `should` and `shouldNot` prompt lists to record what must and must not trigger the skill.7. Run `pnpm check`. Fix every reported `<rule> <path>:<line>` before opening a PR.## Review checklist- The description answers "what", "when" and "not when" in one paragraph.- No secrets, no binaries, no downloads piped into a shell, no instructions that ask the agent to hide information from the user.- Every relative link in `SKILL.md` resolves to a file inside the skill.- `metadata.reviewed` was updated in this change.

Frequently asked questions