Reply Review

What it does

When your agent uses it

• Tasks that involve Pull requests

Example prompts

• “/reply-review”

Requirements

• Python 3
• Pre-approved tools (allowed-tools): Bash(gh api *), Bash(gh pr view *), Bash(gh pr checks *), Bash(git log *), Bash(git show *), Bash(git rev-parse *), Bash(git diff *), Read, Grep

Workflow steps

5 steps, taken from the first numbered list in SKILL.md.

1. 1未解決 thread をリストで表示 (path:line、author、body の要約)
2. 2ユーザーに確認: 「どのコメントに対して、何をコミットで対応したか」
3. 3引数 $2 のハッシュ、または git log dev..HEAD の直近コミットから対応コミットを推定
4. 4コミットの diff (git show --stat ) を確認し、各コメントと修正を紐付け
5. 5紐付け結果をユーザーに提示して承認を得る

What it can do on your machine

Read from SKILL.md and the folder at commit 421d066. It shows what the files ask for, not the result of running them.

• Tool permissions

  Pre-approves these tools, so the agent can use them without asking each time:

  • Bash(gh api *)
  • Bash(gh pr view *)
  • Bash(gh pr checks *)
  • Bash(git log *)
  • Bash(git show *)
  • Bash(git rev-parse *)
  • Bash(git diff *)
  • Read
  • Grep

  From allowed-tools in the SKILL.md frontmatter.

• Runs code

  Shell commands in SKILL.md call:

  • git
  • gh

  From the folder's file list and the shell code blocks in SKILL.md.

• Network

  No URLs in SKILL.md. Its commands use git and gh, which can reach the network depending on how they are called.

  From URLs in SKILL.md, links to its own repository left out.

• Credentials

  Names no API keys, tokens, secrets or passwords.

  From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Reply Review loads about 2.4k tokens when it runs. Until then it costs about 94 tokens; SKILL.md has 556 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it

~94

When it runs · the whole SKILL.md, loaded when a task matches

~2.4k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from ayutaz/piper-plus at commit 421d066, republished under its MIT licence (© ayutaz). 556 words, ~2,354 tokens.

name

reply-review

description

PR レビューコメントへの対応 / review thread の resolve / Copilot や human reviewer のコメントに返信 する文脈で発動。修正コミット後に呼ぶと、各 unresolved thread に対して返信本文を生成し thread を resolve する。`--stale-check` で「コメント以降に該当ファイルが更新済」を自動 flag、`--skip-copilot-style` で Copilot 定型ノイズ regex 除外、`--dry-run` で計画のみ表示、`--auto-safe` で stale + Copilot ノイズ category のみ自動 reply + resolve (人間 reviewer / 非 stale はユーザ判断)。

allowed-tools

Bash(gh api *), Bash(gh pr view *), Bash(gh pr checks *), Bash(git log *), Bash(git show *), Bash(git rev-parse *), Bash(git diff *), Read, Grep

argument-hint

<pr-number> [commit-hash] [--stale-check] [--skip-copilot-style] [--auto-safe] [--dry-run]

disable-model-invocation

false

gh api graphql -f query='
query($pr: Int!) {
  repository(owner: "ayutaz", name: "piper-plus") {
    pullRequest(number: $pr) {
      reviewThreads(first: 50) {
        nodes {
          id
          isResolved
          comments(first: 1) {
            nodes {
              id
              databaseId
              path
              line
              body
              author { login }
              originalCommit { oid }
              commit { oid }
            }
          }
        }
      }
    }
  }
}' -F pr=<PR_NUMBER> --jq '.data.repository.pullRequest.reviewThreads.nodes[] | select(.isResolved == false) | {thread_id: .id, comment: .comments.nodes[0]}'

git log --oneline <originalCommit.oid>..HEAD -- <comment.path> 2>/dev/null

## Stale Check 結果

| # | path:line | author | original_commit | since | 推定 |
|---|-----------|--------|-----------------|-------|------|
| 1 | src/foo.py:42 | Copilot | a7f8abb | 3 commits | 🔶 stale (既に修正済の可能性) |
| 2 | docs/bar.md:10 | reviewer | b2c3d4e | 0 commits | ⏳ 未対応 |

COPILOT_STYLE_NOISE_PATTERNS = [
    r"^Consider (using|renaming|adding|extracting|simplifying)\b",
    r"^Consider whether\b",
    r"\bstyle:\s*prefer\b",
    r"\bsecurity warning\b",
    r"\bMight be worth\b",
    r"^Suggestion: ",
    r"^This (variable|function|class) could be\b",
    r"\b(nit|nit:)\s",
    r"^Optional:",
]

## Copilot Style Noise (除外候補)

| # | path:line | body (先頭 60 文字) | matched pattern |
|---|-----------|-------------------|-----------------|
| 1 | src/foo.py:42 | Consider renaming this variable to ... | ^Consider (renaming) |
| 2 | docs/bar.md:10 | Optional: you could also ...           | ^Optional:        |

これらを除外しますか? (y/N/individual)
- y: 全てスキップ (フェーズ 2 以降の対象から除外)
- N: 通常通り対応
- individual: 1 件ずつ判定

## Auto-safe 自動処理対象 (<N> 件)

| # | path:line | author | category | reply 本文 (40 文字 preview) |
|---|-----------|--------|----------|----------------------------|
| 1 | src/foo.py:42 | Copilot | SAFE-stale | 修正済 (commit a7f8abb)。 該当 file は |
| 2 | docs/bar.md:10 | Copilot | SAFE-copilot-style | style 提案は本 PR の scope 外。 主旨は |

## ユーザ判断必要 (<M> 件)

| # | path:line | author | category | body (60 文字) |
|---|-----------|--------|----------|----------------|
| 1 | src/baz.rs:88 | yousan | REVIEW-human | "logic 順序が逆では?" |
| 2 | src/qux.go:12 | Copilot | REVIEW-copilot-blocker | "TOCTOU race の可能性..." |

対応しました (commit <hash>)。

<修正内容の 1-3 行要約>

<必要なら修正前後のコードスニペット>

<検証結果: テスト PASS カウント等>

gh api repos/ayutaz/piper-plus/pulls/<PR>/comments \
  --method POST \
  -F in_reply_to=<comment_database_id> \
  -f body="$REPLY_BODY" \
  --silent

gh api graphql -f query='
mutation ResolveThread($id: ID!) {
  resolveReviewThread(input: {threadId: $id}) {
    thread { id isResolved }
  }
}' -f id=<THREAD_ID> --jq '.data.resolveReviewThread.thread | "\(.id): resolved=\(.isResolved)"'

## レビュー対応完了

### PR #<N>

| コメント | ファイル | 返信 | Resolve |
|---------|---------|------|---------|
| 1 | src/... | ✅ | ✅ |
| 2 | docs/... | ✅ | ✅ |
| ... | ... | ... | ... |

全 <N> 件 完了。

# PR #349 のレビューに返信+resolve (コミットは HEAD)
/reply-review 349

# PR #350 に対して特定コミットで対応
/reply-review 350 a2e57f05

# Dry-run で計画のみ表示
/reply-review 349 --dry-run

# Stale check 付き: コメント以降のファイル変更を解析して既修正候補を flag
/reply-review 349 --stale-check

# Stale check + dry-run の組み合わせ (検出のみ、投稿しない)
/reply-review 349 --stale-check --dry-run

# Copilot 定型ノイズを除外して対応対象を絞る
/reply-review 349 --skip-copilot-style

# 全フィルタを有効化 (stale 検出 + Copilot ノイズ除外 + dry-run)
/reply-review 349 --stale-check --skip-copilot-style --dry-run

# 自動安全モード: stale + Copilot ノイズだけ自動 reply + resolve、
# 人間 / blocker は user 提示のみ。 watch-pr から chain 起動される。
/reply-review 349 --auto-safe

# 自動安全モード + dry-run: 計画のみ (実際の reply / resolve は走らない)
/reply-review 349 --auto-safe --dry-run

© ayutaz, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in .claude/skills/reply-review of ayutaz/piper-plus.

Open the folder on GitHubat commit 421d066