CH

chatgpt-pro-review

Delegates code review to ChatGPT Pro for external second opinions on SA-AIAPP.

Install

mkdir -p .claude/skills/chatgpt-pro-review && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/13006" && unzip -o skill.zip -d .claude/skills/chatgpt-pro-review && rm skill.zip

Installs to .claude/skills/chatgpt-pro-review

Activation

This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.

当用户要求在 SA-AIAPP 项目中询问 ChatGPT/GPT Pro、让外部模型审核当前 GitHub 分支、做画布/前端/后端/架构二次审查时使用。
78 charsno explicit “when” trigger
Intermediate

Key capabilities

  • Confirm local branch and commit status
  • Open ChatGPT via Browser skill
  • Confirm model is Pro / Pro Advanced before sending review request
  • Prioritize GitHub connector for repository review
  • Send Chinese read-only review request with repository, branch, commit, scope, rules, and format
  • Save question and reply to external consultation directory

How it works

The skill confirms the local branch status, uses the Browser skill to interact with ChatGPT Pro, sends a structured review request, and archives the conversation locally.

Inputs & outputs

You give it
user request to ask ChatGPT/GPT Pro to review the current GitHub branch for SA-AIAPP project
You get back
archived question and reply in the external consultation directory, and a summary to the user

When to use chatgpt-pro-review

  • Requesting architecture review
  • Getting a second opinion on PRs
  • Auditing code patterns with external models
  • Reviewing frontend/backend design

About this skill

ChatGPT Pro Review

Overview

这是 SA-AIAPP 项目级技能。目标是通过已连接的 GitHub 仓库向 ChatGPT Pro 询问第二意见。GPT Pro 是比主 Agent 更强的审查模型;给它清晰目标和边界,让它自主选择查看路径并充分发挥。

Scope

  • 仓库:zhanglongxiao111/SA-AIAPP
  • 默认分支:当前本地分支,通常是 dev
  • 外部咨询归档根目录(本机绝对路径):D:/AI/SA-AIAPP/docs/AI协作/外部咨询/进行中/
  • 外部咨询归档根目录(仓库相对路径):docs/AI协作/外部咨询/进行中/
  • 单次咨询目录:docs/AI协作/外部咨询/进行中/YYYY-MM-DD_<主题>/
  • 优先让它看 GitHub 分支或提交,不优先上传 ZIP。
  • ChatGPT 只提供建议。重要结论必须回到本地代码核实。
  • 不把一次审查的关注点固化成默认流程。按用户当次目标组织提示词。

Workflow

  1. 先确认本地分支和提交:
    • git branch --show-current
    • git rev-parse --short HEAD
    • git status --short --branch
  2. 如果有未提交改动,说明 GPT Pro 看不到,除非先推送或单独贴给它。
  3. 只通过 Browser 技能打开 ChatGPT。用户需要登录时再显式显示浏览器。
  4. 必须确认模型是 Pro / Pro 进阶。如果当前不是 Pro,不要发送审查请求。
  5. 优先使用 GitHub 连接器。只有仓库不可用时才考虑小范围 ZIP。
  6. 发中文只读审查请求。必须包含仓库、分支、提交、目标范围、项目规则和输出格式。
  7. 不打断正在生成的回答。不得点停止、不得追问、不得改写请求;只被动等待。
  8. 回答完成后,先落文档,再向用户总结;并在本地核实具体代码结论后再建议改动。

Documents

每次正式询问 GPT Pro 后,必须把问题和回复保存到外部咨询目录,不能只留在网页里。用户口语说“外部调研目录”时,也按本仓库现有的 docs/AI协作/外部咨询/进行中/ 理解。

目录格式:

D:/AI/SA-AIAPP/docs/AI协作/外部咨询/进行中/YYYY-MM-DD_<主题>/

仓库内相对路径为:

docs/AI协作/外部咨询/进行中/YYYY-MM-DD_<主题>/

文件格式:

提问_<主题>.md
回复_<主题>_GPTPro.md

文档要求:

  • 提问 文件记录背景、目标、仓库分支、提交、实际发送给 GPT Pro 的提示词。
  • 回复 文件记录页面可见的 GPT Pro 结论;如果只整理摘要,必须说明不是完整原文。
  • 回复 文件要写清:这是外部模型意见,后续采纳前必须回到本地代码核实。
  • 中文文档用 UTF-8 写入;写完后用 UTF-8 复读抽查。

Prompt Template

使用中文提示词。避免过拟合,不要把上一次任务的文件名、模块名或结论硬塞进新请求:

请使用已连接的 GitHub 应用审查 `zhanglongxiao111/SA-AIAPP` 仓库,分支 `<branch>`,提交 `<commit>`。

目标:<用户本次要审查的问题或范围>。

请先阅读根目录 `AGENTS.md`,再查看真实代码。你是更强的审查模型,请自主决定应深入查看哪些文件、数据流和边界,不要只按我列出的路径机械搜索。不要修改代码。

请用中文输出:
1. 结论。
2. 实际查看到的关键模块和数据流。
3. 有证据的问题清单,按严重程度排序,并给出文件路径或代码依据。
4. 如果需要调整,给出更优雅的目标架构;如果现状合理,也明确说明哪些部分不该动。
5. 渐进调整计划,避免大爆炸重构。
6. 风险与验证。

不要泛泛而谈。没有证据的判断请标明不确定,或不要写。

如果用户明确要求画布架构审查,可以追加关注点,但不要把它们当成唯一范围:

  • 图片入口中心
  • 程序化场景写入中心
  • 保存协调和回写边界
  • 文件身份 -> 画布身份 -> 图片身份
  • customData 兼容 helper
  • 图片几何中心
  • legacy bridge 和 fallback

Guardrails

  • 不用 Windows 全局按键发送。它会影响用户当前操作。
  • 不把密钥、.env、令牌、本地私有文件发给 ChatGPT。
  • 不把 GPT Pro 当最终真相。本地代码和 AGENTS.md 优先。
  • 用户给了窄范围时,不发散成全仓库审查;用户要求架构判断时,也不要把提示词写成代码搜索清单。
  • 如果 ChatGPT 因请求太大卡住,先等;需要缩小范围时先告诉用户。

When not to use it

  • If there are uncommitted changes that GPT Pro cannot see
  • If the current model is not Pro / Pro Advanced
  • If the task involves sending sensitive local files to ChatGPT

Limitations

  • Only for zhanglongxiao111/SA-AIAPP repository
  • Default branch is current local branch, usually dev
  • ChatGPT only provides suggestions; important conclusions must be verified locally

How it compares

This skill orchestrates an external AI review process for a specific GitHub project, including structured prompting, output archiving, and local verification, which is more formalized than a direct chat with an AI.

Compared to similar skills

chatgpt-pro-review side by side with the closest alternatives in the catalog.

SkillInstallsUpdatedSafetyDifficulty
chatgpt-pro-review (this skill)02moReviewIntermediate
using-serena-for-exploration98moReviewIntermediate
cursor-explorer-mcp68moNo flagsIntermediate
react-expert86moReviewAdvanced

Try saying

Example prompts that trigger this skill in your AI assistant.

You might also like

using-serena-for-exploration

seanGSISG

Use when exploring codebases with Serena MCP tools for architectural understanding and pattern discovery - guides efficient symbolic exploration workflow minimizing token usage through targeted symbol reads, overview tools, and progressive narrowing

9127

cursor-explorer-mcp

sepiabrown

Use for token-expensive operations requiring multi-file analysis - codebase exploration, broad searches, architecture understanding, tracing flows, finding implementations across files. Uses MCP cursor-agent server (company pays) with clean async interface. Do NOT use for single-file analysis, explaining code already in immediate context, or pure reasoning tasks.

699

react-expert

reactjs

Use when researching React APIs or concepts for documentation. Use when you need authoritative usage examples, caveats, warnings, or errors for a React feature.

823

analyzing-projects

CloudAI-X

Analyzes codebases to understand structure, tech stack, patterns, and conventions. Use when onboarding to a new project, exploring unfamiliar code, or when asked "how does this work?" or "what's the architecture?"

327

leann-search

parcadei

Semantic search across codebase using LEANN vector index

29

cartographer

kingbootoshi

Maps and documents codebases of any size by orchestrating parallel subagents. Creates docs/CODEBASE_MAP.md with architecture, file purposes, dependencies, and navigation guides. Updates CLAUDE.md with a summary. Use when user says "map this codebase", "cartographer", "/cartographer", "create codebase map", "document the architecture", "understand this codebase", or when onboarding to a new project. Automatically detects if map exists and updates only changed sections.

37

Search skills

Search the agent skills registry