Tidy Skill

Read-only hygiene audits that stop agents littering repos with plan.md/todo.md junk.

av Phoenix0531-sudo · Phoenix0531-sudo/tidy-skill

Testad · Fungerar ★ 9.6/10

Tidy Skill — Read-only hygiene audits that stop agents littering repos with plan.md/todo.md junk.

Vad den gör

Governs where AI-agent-created files (plans, todos, reports, caches) are allowed to live, and scores a repo's artifact hygiene with a real Python/PowerShell scanner. Triggers on repo cleanup, hygiene scoring, workspace cache audits, and WSL2/Docker/C-drive growth questions; stays out of source code and tool-state directories.

Testrapport

Ran score_repo_hygiene.py on a throwaway repo with plan.md/todo.md/summary.md in root — it returned a real 57/100 'Needs tidy-up' score with a dimension breakdown and named the 3 suspicious files, something a baseline answer only eyeballs.

Testad: 2026-07-16 · Claude Code 2.x (agent harness)

Installation

git clone https://github.com/Phoenix0531-sudo/tidy-skill
mkdir -p ~/.claude/skills
cp -r tidy-skill/skills/tidy-skill ~/.claude/skills/tidy-skill

Kommandon och exempelprompter

/tidy-skillRead-only hygiene audits that stop agents littering repos with plan.md/todo.md junk.

Skills triggas av vanliga förfrågningar — inga kommandon att memorera. Efter installationen aktiverar prompter som dessa skillen (på engelska):

scan my repo root for stray plan.md and todo.md files agents left
score how tidy my workspace is before I hand it to a teammate
why is my WSL2 disk filling up with cache junk from agents?

Läs hela rapporten →