Skip to content
Back to skills

Reports

ASecurity

Critical 0 / Important 0 / Smells 0 - 範圍:worktree-manager/SKILL.md 收尾改寫(diff 檔第二段整檔)及其引用的步驟、輸出碼、`git worktree remove` 行為;未逐檔審其餘 7 個檔案。 沒有達到報告門檻的發現。收尾段的步驟引用(步驟 4 複製 `.env`、步驟 6 寫 `WORKTREE_INFO.md`)與同檔步驟編號一致;`status --porcelain --ignored` 的 ` M`、`A `、`??`、`!!` 描述與 `git worktree remove` 的拒絕/直接刪除行為都與實測相符。 What I ran - `git --version` → git 2.42.0(本機只有這一版,2.54.0 未驗) - 拋棄式 repo(/tmp/wt-exp.2a1Eg8)建 worktree,放 ` M`、`A `、`??`、`!!` 各一類 → `status --porcelain --ignored` 輸出 ` M a.txt` / `A staged.txt` ...

  • 12 stars
  • 0 votes
  • 0 copies
  • 0 views
  • Added November 11, 2026
ai-agentsnodegit

Security analysis

A100/100

Pro scans all 21 files and shows the line behind each finding

Scanned November 11, 2026

npx -y skills add kopp0510/claude-dd --skill reports --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Reports?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Reports
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/kopp0510-reports/badge)](https://www.skillsdirectory.com/skills/kopp0510-reports)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
Critical 0 / Important 0 / Smells 0 - 範圍:worktree-manager/SKILL.md 收尾改寫(diff 檔第二段整檔)及其引用的步驟、輸出碼、`git worktree remove` 行為;未逐檔審其餘 7 個檔案。

沒有達到報告門檻的發現。收尾段的步驟引用(步驟 4 複製 `.env`、步驟 6 寫 `WORKTREE_INFO.md`)與同檔步驟編號一致;`status --porcelain --ignored` 的 ` M`、`A `、`??`、`!!` 描述與 `git worktree remove` 的拒絕/直接刪除行為都與實測相符。

What I ran
- `git --version` → git 2.42.0(本機只有這一版,2.54.0 未驗)
- 拋棄式 repo(/tmp/wt-exp.2a1Eg8)建 worktree,放 ` M`、`A `、`??`、`!!` 各一類 → `status --porcelain --ignored` 輸出 ` M a.txt` / `A  staged.txt` / `?? WORKTREE_INFO.md` / `!! .env` / `!! node_modules/`
- 同一 worktree 不加 `--force` 執行 `git worktree remove` → 拒絕,`致命錯誤: ... 包含修改或未追蹤的檔案,使用 --force 刪除`,exit 128
- 第二個拋棄式 repo(/tmp/wt-exp2.yMjb37)只剩 `!!` 的 `.env` 與 `node_modules/` → `git worktree remove` exit 0,目錄整個消失、無提示(之後的 `No such file` 是預期結果)
- 僅有 staged 新檔 `A ` 的 worktree → `git worktree remove` 拒絕,exit 128
- 讀 diff.txt 第 1644–1699 行(SKILL.md 整檔)與第 1–249 行的 diff,核對步驟編號、CHANGELOG 對應段落

Not reported
- allowed-tools 未列 AskUserQuestion,但建立流程第 2 步已在用,屬既有狀況;根目錄 CLAUDE.md 也說 allowed-tools 只是預授權,最多 minor,故不報。
- 「WORKTREE_INFO.md 每次都會出現在 `??`」是過度概括:若專案 .gitignore 忽略它會變 `!!`,但流程仍正確處理,無失敗情境。
- 「git 2.54.0 實測」無法在本機驗證,需外部查證(`git --version` 2.54.0 環境下重跑上面的 remove 實驗)。
- 收尾步驟 1 的輸出碼列舉不含 ` D`、`R`、`UU` 等,但以「這類」收斂,未造成錯誤行為。
- 新文字取消了舊版「確認 worktree 內變更已全部 commit」的硬性前置,改為交給使用者決定;這是行為變更但與 CHANGELOG 的描述一致,不算缺陷。
- 其餘 7 個檔案(CHANGELOG 除本條外、CLAUDE.md、scripts/CLAUDE.md、verification-gate、templates/global/CLAUDE.md、tests/*)未審,不在本次方向內。

Files in this skill

  • G2-hk-1.d1.aac478.G2-hk-1-judge.md5.7 KB
  • G2-hk-1.d2.a71b72.Reviewer_B__test_scenarios_validity.md2.6 KB
  • G2-hk-1.d2.aa79c1.Reviewer_A__gate_script_logic.md2.4 KB
  • G2-hk-1.d2.ae62d9.Reviewer_C__doc_comment_claims.md3.7 KB
  • G2-hk-2.d1.a6b39a.G2-hk-2-judge.md4.3 KB
  • G2-hk-2.d2.a21bd9.Review_C__doc_claims_vs_code.md2.7 KB
  • G2-hk-2.d2.ab8118.Review_B__new_test_scenarios_validity.md3.5 KB
  • G2-hk-2.d2.aefdca.Review_A__gate_refactor_equivalence.md2.2 KB
  • G2-new-1.d1.a01928.G2-new-1-judge.md7.9 KB
  • G2-new-1.d2.a79b15.Reviewer_C__doc_comment_claims.md4.3 KB
  • G2-new-1.d2.aa94d8.Reviewer_B__test_discrimination.md5.3 KB
  • G2-new-1.d2.abdceb.Reviewer_A__gate_refactor.md3.6 KB
  • G2-new-2.d1.a34615.G2-new-2-judge.md7.3 KB
  • G2-new-2.d2.ab6459.Reviewer_C__doc_comment_claims.md6.1 KB
  • G2-new-2.d2.adae89.Reviewer_B__new_tests_discriminating_power.md3.9 KB
  • G2-new-2.d2.afd492.Reviewer_A__gate_refactor_logic.md3.9 KB
  • G2-old-1.d1.a594bb.G2-old-1-b.md3.9 KB
  • G2-old-1.d1.acef78.G2-old-1-a.md2.4 KB
  • G2-old-2.d1.a0166b.G2-old-2-b.md2.4 KB
  • G2-old-2.d1.a67862.G2-old-2-a.md2.6 KB

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

View all in ai-agents →

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.