From voice-check
Scan a document for AI writing tells, em dashes, hedging, puffery, and tone violations. Use when reviewing or editing prose, when invoked via /voice-check, or before committing any document. Layers global rules with per-repo supplements found at .claude/voice-check.md walking up from the target file.
How this skill is triggered — by the user, by Claude, or both
Slash command
/voice-check:voice-checkThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Scan a file for AI writing patterns and tone violations. Fix every issue found, then report what you changed.
Scan a file for AI writing patterns and tone violations. Fix every issue found, then report what you changed.
../../references/rules.md for the full rule list (the file lives at the plugin root, not inside the skill dir).claude/voice-check.md at each level. Stop at the first one found. Do not aggregate across levels.All reference files live at the plugin root, two levels up from this skill directory:
../../references/rules.md — the rule list (single source of truth)../../references/wikipedia-signs.md — full Wikipedia article on signs of AI writing../../references/examples.md — before/after examples for each rule categoryCreates, edits, and optimizes skills for Claude Code, including drafting, evaluating with test prompts, iterating on performance, and improving skill descriptions for better triggering accuracy.
npx claudepluginhub ryansinger/voice-check --plugin voice-check