Browse our docs https://adamczyk.ovh/docs/skillslint.

Rationale

Agents skills created via Anthropics skill-creator are generally good and reliable. Nevertheless You can also create or edit them manually since they are just SKILL.md files.

To avoid common mistakes, this CLI runs predefined textlint and agent-skills-cli to provide both semantic and quality checks on all Your skills.

Usage

You can install package locally (eg for QoQ usage) or run it directly:

npx @ladamczyk/skillslint

to list available options

npx @ladamczyk/skillslint -h

Anonymous usage stats

Opt-in, off until a human says yes. The first interactive run asks; the answer is stored as stats: true|false in ~/.config/skillslint.json (or $XDG_CONFIG_HOME/skillslint.json) and never asked again — edit or delete the key to change it. Runs that can't ask (CI=true, a pipe) are never prompted and never counted. The JavaScript API never prompts: stats is a required boolean on lint(), so the calling host states its own user's consent.

A counted run posts one constant to https://stats.adamczyk.ovh and nothing else — every run, every flag combination, byte for byte the same body:

{ "tool": "skillslint", "options": [] } // `options` is always empty

Where an outbound POST never leaves the network, the same run counts as a plain image GET instead — https://adamczyk.ovh/img/stats/pixel.png?tool=skillslint — which carries the same single value and nothing more.

Never sent: your skills, file names, paths, scores, findings, thresholds, the flags you typed, project or package names, or anything identifying the user or machine. Sends are fire-and-forget with a 2s timeout; a failure is swallowed and never affects the exit code.

Last 12 months usage statistics (>=4.2.0)

Regenerated on each release, for daily updated numbers click on chart or visit https://adamczyk.ovh/docs/skillslint#usage-statistics.