Back to Details

create-review

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

Create Review

生成AI代码审查提示

Bootstrap the review pipeline and generate a paste-ready review prompt for any AI reviewer.
搭建审查流水线，并为任意AI审查工具生成可直接粘贴使用的审查提示词。

Step 1: Determine the reviewer and session

步骤1：确定审查工具与会话

Reviewer: If the user provided 
$ARGUMENTS
, sanitize it to a safe kebab-case string (lowercase, strip any characters that aren't alphanumeric or hyphens, collapse multiple hyphens) and use that as the reviewer name (e.g. 
codex
, 
gemini
, 
gpt4
, 
claude
). If no argument, use 
generic
.
Session: Detect from context:
  1. If 
    .chalk/reviews/
    exists, check for the most recent session directory
  2. Otherwise, infer from the current branch name (kebab-case)
  3. If on 
    main
    /
    master
    , ask the user
Store as 
{reviewer}
and 
{session}
.
审查工具： 如果用户提供了
$ARGUMENTS
，将其处理为安全的短横线分隔字符串（小写，移除所有非字母数字或短横线的字符，合并多个短横线），并将其作为审查工具名称（例如
codex
、
gemini
、
gpt4
、
claude
）。如果没有提供参数，则使用
generic
。
会话： 从上下文检测：
  1. 如果存在
    .chalk/reviews/
    目录，检查最新的会话目录
  2. 否则，从当前分支名称推断（转换为短横线分隔格式）
  3. 如果当前分支是
    main
    /
    master
    ，询问用户
将结果存储为
{reviewer}
和
{session}
。

Step 2: Bootstrap the review pipeline

步骤2：搭建审查流水线

Check if 
.chalk/reviews/scripts/pack.sh
exists. If not, bootstrap the full pipeline:
sh
mkdir -p .chalk/reviews/scripts .chalk/reviews/templates .chalk/reviews/sessions
检查是否存在
.chalk/reviews/scripts/pack.sh
。如果不存在，则搭建完整流水线：
sh
mkdir -p .chalk/reviews/scripts .chalk/reviews/templates .chalk/reviews/sessions

Create 
.chalk/reviews/scripts/pack.sh

创建
.chalk/reviews/scripts/pack.sh
脚本

This script generates a review context pack from git state:
sh
#!/usr/bin/env bash
set -euo pipefail

BASE_REF="${1:-origin/main}"
SESSION="${2:-adhoc}"
OUTPUT_PATH="${3:-.chalk/reviews/sessions/${SESSION}/pack.md}"
该脚本从Git状态生成审查上下文包：
sh
#!/usr/bin/env bash
set -euo pipefail

BASE_REF="${1:-origin/main}"
SESSION="${2:-adhoc}"
OUTPUT_PATH="${3:-.chalk/reviews/sessions/${SESSION}/pack.md}"

Resolve base ref

Resolve base ref

if ! git rev-parse --verify "$BASE_REF" >/dev/null 2>&1; then for candidate in main origin/main master origin/master; do if git rev-parse --verify "$candidate" >/dev/null 2>&1; then BASE_REF="$candidate" break fi done fi
MERGE_BASE="$(git merge-base HEAD "$BASE_REF" 2>/dev/null || echo "")" if [ -z "$MERGE_BASE" ]; then MERGE_BASE="$(git rev-list --max-parents=0 HEAD | tail -n 1)" fi
mkdir -p "$(dirname "$OUTPUT_PATH")"
{ echo "# Review Pack" echo echo "- Session: `$SESSION`" echo "- Generated: $(date -u +"%Y-%m-%d %H:%M:%S UTC")" echo "- Base ref: `$BASE_REF`" echo "- Merge base: `${MERGE_BASE:0:12}`" echo "- Head: `$(git rev-parse --short HEAD)`" echo echo "## Diff Stat" echo '
'   git diff --stat "$MERGE_BASE"..HEAD 2>/dev/null || echo "(no committed diff)"   echo '
' echo echo "## Changed Files" CHANGED="$(git diff --name-only "$MERGE_BASE"..HEAD 2>/dev/null || true)" if [ -n "$CHANGED" ]; then echo "$CHANGED" | while IFS= read -r f; do [ -n "$f" ] && echo "- $f"; done else echo "- (none)" fi echo echo "## Commit Log" echo '
'   git log --oneline "$MERGE_BASE"..HEAD 2>/dev/null || echo "(no commits ahead of base)"   echo '
' echo echo "## Working Tree Status" echo '
'   git status --short   echo '
' } > "$OUTPUT_PATH"
echo "PACK_PATH=$OUTPUT_PATH"
undefined
if ! git rev-parse --verify "$BASE_REF" >/dev/null 2>&1; then for candidate in main origin/main master origin/master; do if git rev-parse --verify "$candidate" >/dev/null 2>&1; then BASE_REF="$candidate" break fi done fi
MERGE_BASE="$(git merge-base HEAD "$BASE_REF" 2>/dev/null || echo "")" if [ -z "$MERGE_BASE" ]; then MERGE_BASE="$(git rev-list --max-parents=0 HEAD | tail -n 1)" fi
mkdir -p "$(dirname "$OUTPUT_PATH")"
{ echo "# Review Pack" echo echo "- Session: `$SESSION`" echo "- Generated: $(date -u +"%Y-%m-%d %H:%M:%S UTC")" echo "- Base ref: `$BASE_REF`" echo "- Merge base: `${MERGE_BASE:0:12}`" echo "- Head: `$(git rev-parse --short HEAD)`" echo echo "## Diff Stat" echo '
'   git diff --stat "$MERGE_BASE"..HEAD 2>/dev/null || echo "(no committed diff)"   echo '
' echo echo "## Changed Files" CHANGED="$(git diff --name-only "$MERGE_BASE"..HEAD 2>/dev/null || true)" if [ -n "$CHANGED" ]; then echo "$CHANGED" | while IFS= read -r f; do [ -n "$f" ] && echo "- $f"; done else echo "- (none)" fi echo echo "## Commit Log" echo '
'   git log --oneline "$MERGE_BASE"..HEAD 2>/dev/null || echo "(no commits ahead of base)"   echo '
' echo echo "## Working Tree Status" echo '
'   git status --short   echo '
' } > "$OUTPUT_PATH"
echo "PACK_PATH=$OUTPUT_PATH"
undefined

Create 
.chalk/reviews/scripts/render-prompt.sh

创建
.chalk/reviews/scripts/render-prompt.sh
脚本

This script combines pack + handoff + reviewer template into a prompt:
sh
#!/usr/bin/env bash
set -euo pipefail

REVIEWER="${1:?Usage: render-prompt.sh <reviewer> [pack-path] [handoff-path] [output-path]}"
PACK_PATH="${2:-}"
HANDOFF_PATH="${3:-}"
OUTPUT_PATH="${4:-}"
SESSION="${5:-}"
SCRIPT_DIR="$(cd "$(dirname "$0")" && pwd)"
ROOT_DIR="$(dirname "$SCRIPT_DIR")"

REVIEWER_TITLE="$(echo "$REVIEWER" | awk '{print toupper(substr($0,1,1)) substr($0,2)}')"

if [ -z "$SESSION" ] && [ -f "$ROOT_DIR/.current-session" ]; then
  SESSION="$(cat "$ROOT_DIR/.current-session")"
fi
SESSION="${SESSION:-adhoc}"

[ -z "$PACK_PATH" ] && PACK_PATH="$ROOT_DIR/sessions/$SESSION/pack.md"
[ -z "$HANDOFF_PATH" ] && HANDOFF_PATH="$ROOT_DIR/sessions/$SESSION/handoff.md"
[ -z "$OUTPUT_PATH" ] && OUTPUT_PATH="$ROOT_DIR/sessions/$SESSION/${REVIEWER}.prompt.md"

if [ ! -f "$PACK_PATH" ]; then
  echo "Pack not found at $PACK_PATH. Run pack.sh first." >&2
  exit 1
fi

mkdir -p "$(dirname "$OUTPUT_PATH")"

{
  echo "# $REVIEWER_TITLE Review Request"
  echo

  # Use provider-specific template if available, else generic
  TEMPLATE="$ROOT_DIR/templates/${REVIEWER}-review.template.md"
  if [ ! -f "$TEMPLATE" ]; then
    TEMPLATE="$ROOT_DIR/templates/generic-review.template.md"
  fi

  if [ -f "$TEMPLATE" ]; then
    cat "$TEMPLATE"
  fi

  echo
  echo "---"
  echo
  echo "## Review Pack"
  cat "$PACK_PATH"

  if [ -f "$HANDOFF_PATH" ]; then
    echo
    echo "---"
    echo
    echo "## Handoff"
    cat "$HANDOFF_PATH"
  fi
} > "$OUTPUT_PATH"

echo "PROMPT_PATH=$OUTPUT_PATH"
该脚本将审查包、交接文档和审查工具模板合并为提示词：
sh
#!/usr/bin/env bash
set -euo pipefail

REVIEWER="${1:?Usage: render-prompt.sh <reviewer> [pack-path] [handoff-path] [output-path]}"
PACK_PATH="${2:-}"
HANDOFF_PATH="${3:-}"
OUTPUT_PATH="${4:-}"
SESSION="${5:-}"
SCRIPT_DIR="$(cd "$(dirname "$0")" && pwd)"
ROOT_DIR="$(dirname "$SCRIPT_DIR")"

REVIEWER_TITLE="$(echo "$REVIEWER" | awk '{print toupper(substr($0,1,1)) substr($0,2)}')"

if [ -z "$SESSION" ] && [ -f "$ROOT_DIR/.current-session" ]; then
  SESSION="$(cat "$ROOT_DIR/.current-session")"
fi
SESSION="${SESSION:-adhoc}"

[ -z "$PACK_PATH" ] && PACK_PATH="$ROOT_DIR/sessions/$SESSION/pack.md"
[ -z "$HANDOFF_PATH" ] && HANDOFF_PATH="$ROOT_DIR/sessions/$SESSION/handoff.md"
[ -z "$OUTPUT_PATH" ] && OUTPUT_PATH="$ROOT_DIR/sessions/$SESSION/${REVIEWER}.prompt.md"

if [ ! -f "$PACK_PATH" ]; then
  echo "Pack not found at $PACK_PATH. Run pack.sh first." >&2
  exit 1
fi

mkdir -p "$(dirname "$OUTPUT_PATH")"

{
  echo "# $REVIEWER_TITLE Review Request"
  echo

  # Use provider-specific template if available, else generic
  TEMPLATE="$ROOT_DIR/templates/${REVIEWER}-review.template.md"
  if [ ! -f "$TEMPLATE" ]; then
    TEMPLATE="$ROOT_DIR/templates/generic-review.template.md"
  fi

  if [ -f "$TEMPLATE" ]; then
    cat "$TEMPLATE"
  fi

  echo
  echo "---"
  echo
  echo "## Review Pack"
  cat "$PACK_PATH"

  if [ -f "$HANDOFF_PATH" ]; then
    echo
    echo "---"
    echo
    echo "## Handoff"
    cat "$HANDOFF_PATH"
  fi
} > "$OUTPUT_PATH"

echo "PROMPT_PATH=$OUTPUT_PATH"

Create 
.chalk/reviews/scripts/copy-prompt.sh

创建
.chalk/reviews/scripts/copy-prompt.sh
脚本

sh
#!/usr/bin/env bash
set -euo pipefail

REVIEWER="${1:?Usage: copy-prompt.sh <reviewer> [pack] [handoff] [output] [session]}"
SCRIPT_DIR="$(cd "$(dirname "$0")" && pwd)"

OUTPUT="$(bash "$SCRIPT_DIR/render-prompt.sh" "$@")"
echo "$OUTPUT"

PROMPT_PATH="$(echo "$OUTPUT" | sed -n 's/^PROMPT_PATH=//p' | head -1)"
if [ -z "$PROMPT_PATH" ] || [ ! -f "$PROMPT_PATH" ]; then
  echo "Could not resolve prompt path." >&2
  exit 1
fi

COPIED=0
if command -v pbcopy >/dev/null 2>&1; then
  pbcopy < "$PROMPT_PATH" && COPIED=1 && echo "CLIPBOARD=pbcopy"
elif command -v xclip >/dev/null 2>&1; then
  xclip -selection clipboard < "$PROMPT_PATH" && COPIED=1 && echo "CLIPBOARD=xclip"
elif command -v wl-copy >/dev/null 2>&1; then
  wl-copy < "$PROMPT_PATH" && COPIED=1 && echo "CLIPBOARD=wl-copy"
fi

if [ "$COPIED" -eq 0 ]; then
  echo "CLIPBOARD=none (copy manually from $PROMPT_PATH)"
fi
sh
#!/usr/bin/env bash
set -euo pipefail

REVIEWER="${1:?Usage: copy-prompt.sh <reviewer> [pack] [handoff] [output] [session]}"
SCRIPT_DIR="$(cd "$(dirname "$0")" && pwd)"

OUTPUT="$(bash "$SCRIPT_DIR/render-prompt.sh" "$@")"
echo "$OUTPUT"

PROMPT_PATH="$(echo "$OUTPUT" | sed -n 's/^PROMPT_PATH=//p' | head -1)"
if [ -z "$PROMPT_PATH" ] || [ ! -f "$PROMPT_PATH" ]; then
  echo "Could not resolve prompt path." >&2
  exit 1
fi

COPIED=0
if command -v pbcopy >/dev/null 2>&1; then
  pbcopy < "$PROMPT_PATH" && COPIED=1 && echo "CLIPBOARD=pbcopy"
elif command -v xclip >/dev/null 2>&1; then
  xclip -selection clipboard < "$PROMPT_PATH" && COPIED=1 && echo "CLIPBOARD=xclip"
elif command -v wl-copy >/dev/null 2>&1; then
  wl-copy < "$PROMPT_PATH" && COPIED=1 && echo "CLIPBOARD=wl-copy"
fi

if [ "$COPIED" -eq 0 ]; then
  echo "CLIPBOARD=none (copy manually from $PROMPT_PATH)"
fi

Create 
.chalk/reviews/templates/generic-review.template.md

创建
.chalk/reviews/templates/generic-review.template.md
模板

Only create if it does not already exist (preserve user customizations):
markdown
You are acting as an independent code reviewer.

Task:
- Review the diff only.
- Report defects and risks, not style preferences.

Output format:
1. Findings (highest severity first)
2. Open questions/assumptions
3. Residual risks/testing gaps

Finding schema:
- Severity: P0 (critical) | P1 (high) | P2 (medium) | P3 (low)
- File: <path>:<line>
- Issue: concise summary
- Failure mode: what breaks and when
- Suggested fix: actionable next step

Rules:
- Do not suggest broad refactors unless required for correctness.
- If no blocking issues exist, explicitly state: `No blocking findings`.
仅当模板不存在时创建（保留用户自定义内容）：
markdown
你将作为独立代码审查员开展工作。

任务：
- 仅审查代码差异。
- 报告缺陷和风险，而非风格偏好。

输出格式：
1. 发现问题（按严重程度从高到低排序）
2. 待解决问题/假设
3. 剩余风险/测试缺口

问题描述格式：
- 严重程度：P0（关键）| P1（高）| P2（中）| P3（低）
- 文件：<路径>:<行号>
- 问题：简洁总结
- 失效模式：何时、何处会出现问题
- 建议修复方案：可执行的下一步操作

规则：
- 除非为了确保正确性，否则不建议大范围重构。
- 如果没有阻塞性问题，请明确说明：`No blocking findings`。

Create 
.chalk/reviews/templates/codex-review.template.md

创建
.chalk/reviews/templates/codex-review.template.md
模板

Only create if it does not already exist:
markdown
You are Codex performing an independent PR review.

Primary objective:
- Find real defects and risks in changed lines only.
- Prioritize actionable, high-signal output over style commentary.

Output format (required):
1. `## Verdict`
   - `Block merge: yes|no`
   - `Blocking findings: P0=<n>, P1=<n>`
   - If no P0/P1 findings, include exact text: `No blocking findings`.
2. `## Findings`
   - Markdown table: `ID | Severity | File:Line | Issue | Failure mode | Recommended fix | Confidence`
   - IDs: F-001, F-002, ...
3. `## Testing Gaps`
   - Missing tests that could hide regressions.
4. `## Open Questions`
   - Only unresolved assumptions that affect correctness.

Rules:
- Focus on correctness, security, reliability, and regression risk.
- Do not comment on formatting/import ordering/trivial naming.
- Keep recommendations patch-oriented and specific.
仅当模板不存在时创建：
markdown
你是Codex，正在执行独立的PR审查。

主要目标：
- 仅在修改的代码行中查找真实缺陷和风险。
- 优先输出可执行、高价值的内容，而非风格评论。

输出格式（必填）：
1. `## 评审结论`
   - `是否阻止合并：是|否`
   - `阻塞性问题数量：P0=<n>, P1=<n>`
   - 如果没有P0/P1级问题，请包含确切文本：`No blocking findings`。
2. `## 发现问题`
   - Markdown表格：`ID | 严重程度 | 文件:行号 | 问题 | 失效模式 | 建议修复方案 | 置信度`
   - ID格式：F-001, F-002, ...
3. `## 测试缺口`
   - 可能隐藏回归问题的缺失测试。
4. `## 待解决问题`
   - 仅包含影响正确性的未解决假设。

规则：
- 聚焦于正确性、安全性、可靠性和回归风险。
- 不评论格式/导入顺序/命名等细节问题。
- 建议的修复方案需具体且针对代码补丁。

Create 
.chalk/reviews/templates/gemini-review.template.md

创建
.chalk/reviews/templates/gemini-review.template.md
模板

Only create if it does not already exist:
markdown
You are Gemini performing an independent PR review.

Primary objective:
- Provide a rigorous risk assessment on changed code.
- Emphasize user impact and merge risk.

Output format (required):
1. `## Executive Summary`
   - 2-4 sentences on overall risk.
   - `Merge recommendation: approve|changes-requested`
   - If no P0/P1 findings, include exact text: `No blocking findings`.
2. `## Findings`
   - Markdown table: `ID | Severity | Category | File:Line | Issue | Impact | Suggested fix | Confidence`
   - IDs: G-001, G-002, ...
   - Categories: Security | Correctness | Performance | Reliability | Testing
3. `## Regression & Testing Gaps`
   - Missing test coverage and risky edge cases.
4. `## Assumptions`
   - Only assumptions that materially affect conclusions.

Rules:
- Review changed lines only.
- No style-only feedback.
- Suggested fixes must be concrete and immediately actionable.
仅当模板不存在时创建：
markdown
你是Gemini，正在执行独立的PR审查。

主要目标：
- 对修改后的代码进行严格的风险评估。
- 强调用户影响和合并风险。

输出格式（必填）：
1. `## 执行摘要`
   - 2-4句话总结整体风险。
   - `合并建议：批准|需要修改`
   - 如果没有P0/P1级问题，请包含确切文本：`No blocking findings`。
2. `## 发现问题`
   - Markdown表格：`ID | 严重程度 | 类别 | 文件:行号 | 问题 | 影响 | 建议修复方案 | 置信度`
   - ID格式：G-001, G-002, ...
   - 类别：安全 | 正确性 | 性能 | 可靠性 | 测试
3. `## 回归与测试缺口`
   - 缺失的测试覆盖和高风险边缘场景。
4. `## 假设条件`
   - 仅包含对结论有重大影响的假设。

规则：
- 仅审查修改的代码行。
- 不提供仅涉及风格的反馈。
- 建议的修复方案需具体且可立即执行。

Make scripts executable

为脚本添加执行权限

sh
chmod +x .chalk/reviews/scripts/pack.sh .chalk/reviews/scripts/render-prompt.sh .chalk/reviews/scripts/copy-prompt.sh
sh
chmod +x .chalk/reviews/scripts/pack.sh .chalk/reviews/scripts/render-prompt.sh .chalk/reviews/scripts/copy-prompt.sh

Create 
.chalk/reviews/PIPELINE.md

创建
.chalk/reviews/PIPELINE.md
文档

Write a brief usage guide explaining the pipeline, available scripts, and how to add custom reviewer templates. Refresh this on every run.
编写简要使用指南，说明流水线的使用方法、可用脚本以及如何添加自定义审查工具模板。每次运行时更新该文档。

Step 3: Resolve the base branch

步骤3：确定基准分支

  1. git merge-base main HEAD
    → if it works, use it
  2. Try 
    origin/main
    , then 
    master
    , 
    origin/master
  3. Store as 
    {base}
  1. 执行
    git merge-base main HEAD
    → 如果成功，使用该结果
  2. 尝试
    origin/main
    ，然后是
    master
    、
    origin/master
  3. 将结果存储为
    {base}

Step 4: Check for a handoff

步骤4：检查交接文档

Look for 
.chalk/reviews/sessions/{session}/handoff.md
. If it exists, it will be included in the prompt. If not, warn the user that no handoff was found and suggest running 
/create-handoff
first, but continue anyway.
查找
.chalk/reviews/sessions/{session}/handoff.md
文件。如果存在，将其包含在提示词中。如果不存在，提醒用户未找到交接文档，并建议先运行
/create-handoff
，但仍继续执行流程。

Step 5: Generate the review pack

步骤5：生成审查包

sh
bash .chalk/reviews/scripts/pack.sh "{base}" "{session}" ".chalk/reviews/sessions/{session}/pack.md"
sh
bash .chalk/reviews/scripts/pack.sh "{base}" "{session}" ".chalk/reviews/sessions/{session}/pack.md"

Step 6: Generate the review prompt

步骤6：生成审查提示词

sh
bash .chalk/reviews/scripts/render-prompt.sh "{reviewer}" \
  ".chalk/reviews/sessions/{session}/pack.md" \
  ".chalk/reviews/sessions/{session}/handoff.md" \
  ".chalk/reviews/sessions/{session}/{reviewer}.prompt.md" \
  "{session}"
sh
bash .chalk/reviews/scripts/render-prompt.sh "{reviewer}" \
  ".chalk/reviews/sessions/{session}/pack.md" \
  ".chalk/reviews/sessions/{session}/handoff.md" \
  ".chalk/reviews/sessions/{session}/{reviewer}.prompt.md" \
  "{session}"

Step 7: Copy to clipboard

步骤7：复制到剪贴板

sh
bash .chalk/reviews/scripts/copy-prompt.sh "{reviewer}" \
  ".chalk/reviews/sessions/{session}/pack.md" \
  ".chalk/reviews/sessions/{session}/handoff.md" \
  ".chalk/reviews/sessions/{session}/{reviewer}.prompt.md" \
  "{session}"
sh
bash .chalk/reviews/scripts/copy-prompt.sh "{reviewer}" \
  ".chalk/reviews/sessions/{session}/pack.md" \
  ".chalk/reviews/sessions/{session}/handoff.md" \
  ".chalk/reviews/sessions/{session}/{reviewer}.prompt.md" \
  "{session}"

Step 8: Report to the user

步骤8：向用户反馈结果

Show:
  • The prompt file path
  • Whether it was copied to clipboard
  • The reviewer template used (provider-specific or generic)
  • Suggest: paste the prompt into the target model, or run with a different provider (e.g. 
    /create-review gemini
    )
Also mention:
  • To add a custom reviewer, create 
    .chalk/reviews/templates/{name}-review.template.md
  • To review with multiple providers: run the skill again with a different argument
展示以下信息：
  • 提示词文件路径
  • 是否已复制到剪贴板
  • 使用的审查工具模板（提供商专属或通用模板）
  • 建议：将提示词粘贴到目标AI模型中，或使用其他提供商重新运行（例如
    /create-review gemini
    ）
同时说明：
  • 要添加自定义审查工具，请创建
    .chalk/reviews/templates/{name}-review.template.md
    模板文件
  • 要使用多个提供商进行审查：使用不同参数重新运行该工具

Step 9: Save current session

步骤9：保存当前会话

Write the session name to 
.chalk/reviews/.current-session
so subsequent runs can pick it up.
将会话名称写入
.chalk/reviews/.current-session
文件，以便后续运行时可以直接使用该会话。

Rules

规则

  • Only create template files if they don't already exist — preserve user customizations
  • Always refresh scripts (pack.sh, render-prompt.sh, copy-prompt.sh) to latest version
  • Always refresh PIPELINE.md to latest version
  • Do NOT modify any source code
  • If no git changes exist ahead of base, warn the user but still generate (they may have local uncommitted work)
  • 仅当模板文件不存在时才创建——保留用户自定义内容
  • 始终将脚本（pack.sh、render-prompt.sh、copy-prompt.sh）更新至最新版本
  • 始终将PIPELINE.md更新至最新版本
  • 不得修改任何源代码
  • 如果基准分支之后没有Git变更，提醒用户但仍继续生成（用户可能有本地未提交的修改）