メニュー
Use to audit or clean Rudder worktrees, generated artifacts, logs, caches, and repo-owned processes without deleting active work, user data, or unrelated machine state. Returns AUDIT, CLEANED, or BLOCKED.
デフォルトでは、最初にソースを確認する Prompt が選択されています。直接コマンドに切り替えるか、ローカルコピーをダウンロードすることもできます。
インストールを決める前に、SKILL.md と SkillsMP に表示されている付属ファイルをお読みください。
SOC 職業分類に基づく
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
直接コマンドでは確認用 Prompt が省略されます。実行前にソースを確認してください。
npx skills add https://github.com/Undertone0809/rudder --skill rudder-workspace-hygiene-maintainerコマンドは1行のまま表示されます。コピー前に横へスクロールして全体を確認してください。
ローカルで確認しますか？SkillsMP が現在取得できるファイルをダウンロードできます。
SKILL.md を表示中
Use when the user asks how Rudder works; needs current Rudder docs, exact behavior, commands, contracts, APIs, or source evidence; asks to create, hire, or configure Rudder agents; or asks to scaffold, develop, or verify Rudder plugins. For explicit creation requests, verify the current interface before acting within existing authorization. Do not use for greetings, ordinary work merely running inside Rudder, routine actions already clear from the active context and typed tools, non-Rudder plugin work, or general coding and research tasks that do not ask about Rudder.
Use when inspecting, preparing, executing, recovering, or verifying Rudder releases across npm, GitHub Releases, Desktop assets, tags, dist-tags, changelogs, install smoke, stable/canary promotion, rollback, and release workflow failures. Use for both hands-on publish requests and read-only release readiness questions; route to the smallest relevant release reference instead of loading every release branch.
Use for independent review of Rudder implementations, UI workflows, proposals, agent work, or final handoffs. Judges first-principles intent, raw-request and correction adherence, functional trust, adversarial risk, product taste, and evidence integrity; requires traceable acceptance-packet alignment and comparative rendered evidence, and blocks packet mismatch. Returns accept, conditional accept, needs more evidence, or reject without implementing fixes.
| name | rudder-workspace-hygiene-maintainer |
| description | Use to audit or clean Rudder worktrees, generated artifacts, logs, caches, and repo-owned processes without deleting active work, user data, or unrelated machine state. Returns AUDIT, CLEANED, or BLOCKED. |
Make workspace cost visible and reclaim only space that is both proven safe and explicitly authorized.
AUDIT: read-only inventory and classified cleanup candidates.CLEANED: exact authorized targets removed, with reclaimed space and
post-cleanup health reported.BLOCKED: safe classification or required cleanup authority is missing.An audit is always allowed. Cleanup is allowed only for exact targets covered by the user's request.
Never delete or rewrite the main checkout; dirty, unmerged, locked, or active
worktrees; source or user-created data; ~/.rudder instances and databases;
global caches; unrelated project state; or logs still being written before
identifying and handling their producer.
Do not use broad unresolved globs, inferred environment variables, or recursive deletion against a workspace root.
Run the bundled read-only helper first:
.agents/skills/maintainer/rudder-workspace-hygiene-maintainer/scripts/audit-workspace.sh
Add --sizes only when a slower serial disk scan is worthwhile.
Gross directory size is not reclaimable space. For every candidate, classify: registered worktree and branch; main/current/locked state; dirty/unmerged state; active repo-owned process; merge ancestry or other recoverability proof; generated versus user data; and exact expected reclaim.
Process reports must be scoped to registered Rudder worktree roots or an exact Rudder-owned signature. Report only redacted PID, PPID, elapsed time, RSS, executable, and owner; never dump full argv that may contain tokens.
For worktrees, “old” or “large” is insufficient. Protect any worktree that is dirty, unmerged, active, locked, current, or not proven recoverable from a remote/merged commit.
RESULT: AUDIT | CLEANED | BLOCKED
Scope:
Protected:
Candidates:
Actions:
Reclaimed:
Post-check: