New skill

Skill soonjune/skills/skills/new-skill

Scaffold a new skill in this skills repository — copies the template, enforces Agent Skills spec conventions, refreshes the catalog and marketplace manifest, and relinks agent skill directories. Use when asked to create, add, or scaffold a skill (새 스킬, 스킬 만들어줘, add a skill).From its SKILL.md

Install

npx -y skills add soonjune/skills --skill new-skill

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its file declares

Copied from the file, not written here

The file declares its own license as MIT. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.

SKILL.md

1.8 KB, 342 tokens by cl100k_base, as published. Nobody here has run it

New Skill

Create a skill following this repository's conventions. Skills here serve multiple agents (Claude Code, OpenClaw, Hermes) through the Agent Skills open standard, wired up by symlinks.

Steps

Name it. Lowercase letters, digits, hyphens; max 64 chars; no leading/trailing/consecutive hyphens. The directory name must equal the frontmatter name.
Scaffold. From the repo root: cp -r template skills/<name>, then edit skills/<name>/SKILL.md.
Write the description (max 1024 chars, one line, third person): what it does AND when to trigger it, with the concrete phrases a request would contain, in every language the owner actually uses.
Write the body. Imperative instructions, under 500 lines. Long reference material goes to references/*.md with explicit "load this when..." pointers; runnable helpers go to scripts/.
Cross-agent rules. Shared skills stick to standard frontmatter fields (name, description, license, compatibility, metadata). To restrict a skill to some agents, set metadata.agents: claude openclaw hermes (space-separated subset).

Validate and wire up from the repo root:

python3 scripts/validate.py && python3 scripts/catalog.py
scripts/link.sh

Commit. Never put secrets in skill text — skill files are injected into model context; reference credential file paths or env var names instead.

What ships with it

Read from the repository

Just SKILL.md. No reference files, no scripts.