重要前提
安装AI Skills的关键前提是:必须科学上网,且开启TUN模式,这一点至关重要,直接决定安装能否顺利完成,在此郑重提醒三遍:科学上网,科学上网,科学上网。查看完整安装教程 →
plan-execute by longranger2/claude-gpt-workflow
npx skills add https://github.com/longranger2/claude-gpt-workflow --skill plan-execute当用户运行 /plan-execute {plan-file-path} 时,启动“编排式计划执行”工作流:
reviews/ 目录,然后要求 Codex 检查并修复问题。核心原则:我本人不编写或编辑代码。我只做两件事:审查代码和编排 Codex。所有代码变更,包括实现和修复,都由 Codex 执行。
/plan-execute plans/my-feature-plan.md
每次调用 Codex 后,从脚本输出中提取 session_id=xxx 并保存为当前任务的会话 ID。在后续针对同一任务的 Codex 调用中,传递 --session <id> 以复用上下文,这样 Codex 就能记住之前的实现和修复历史,而不是每次都重新读取整个代码库。
阅读指定的计划文件并理解:
广告位招租
在这里展示您的产品或服务
触达数万 AI 开发者,精准高效
CLAUDE.md 的约定如果计划已经包含一个检查清单(- [ ] / - [x]),则使用这些项作为执行单元。如果计划没有定义明确的步骤,则将工作拆分为合理的批次,每批次修改的文件不超过 5 个。
使用 /codex 技能,并给 Codex 以下指令:
Implement the code according to the plan in {plan-file-path}.
Current execution scope: {specific step or batch description}
Requirements:
- Follow the design in the plan exactly. Do not improvise beyond it.
- Obey the Code Quality Hard Limits defined in `CLAUDE.md`.
- Single file <= 800 lines, single function <= 50 lines, nesting <= 3 levels
- Run `pnpm build` after implementation to confirm compilation succeeds
- If the plan includes a checklist, mark completed steps as `[x]`
After implementation, list all changed files and provide a summary of each change.
Codex 完成后,我执行代码审查。重要:我只阅读代码和撰写审查意见。我从不直接修改源文件。
pnpm build 以确认编译状态。将审查意见追加到 reviews/{topic}-review.md(与 plan-review 共享):
---
## Code Review Round {N} — {YYYY-MM-DD}
**Scope**: {code scope covered in this review}
**Build Status**: PASS / FAIL
### Issues
#### Issue 1 ({severity}): {title}
**File**: {file-path:line}
{issue description}
**Fix**: {specific fix recommendation}
...
### Verdict: NEEDS_FIX / APPROVED
如果 Verdict: NEEDS_FIX,则调用 /codex 让 Codex 修复问题,而不是我自己编辑:
Read the latest Code Review round in {review-file-path}.
Check each issue one by one. Fix the valid issues, and explain why any disputed item is not actually a problem.
After making fixes, run `pnpm build` to confirm compilation succeeds.
List the issues that were fixed and the corresponding code changes.
如果 Verdict: APPROVED,则跳到步骤 6。
在 Codex 应用修复后,我再次审查,仍然不直接编辑代码:
Verdict: APPROVED每批次完成后,要求 Codex 更新计划文件中的检查清单(- [ ] -> - [x])。如果还有未完成的步骤,则返回步骤 2 处理下一批次。一旦所有工作完成,则进入收尾阶段。
向用户报告以下内容:
| 级别 | 含义 | 必须修复 |
|---|---|---|
| 严重 | 导致运行时故障或安全漏洞 | 是 |
| 高 | 违反项目约定或存在明显的设计缺陷 | 是 |
| 中 | 应改进的代码质量问题 | 建议 |
| 低 | 风格或偏好问题 | 可选 |
| 建议 | 优化建议 | 可选 |
裁决规则:
NEEDS_FIXAPPROVED,可附带可选的改进说明plan-review 共享同一个审查文件:reviews/{topic}-review.md{topic} 是计划文件名(不带 .md 后缀)## Round {N},代码审查用 ## Code Review Round {N}每周安装次数
64
仓库
GitHub 星标数
43
首次出现
12 天前
安全审计
安装于
codex63
gemini-cli61
claude-code61
github-copilot59
opencode59
cline58
When the user runs /plan-execute {plan-file-path}, start the "orchestrated plan execution" workflow:
reviews/ directory, then ask Codex to inspect and fix the issues.Core principle: I do not write or edit code myself. I only do two things: review code and orchestrate Codex. All code changes, including implementation and fixes, are performed by Codex.
/plan-execute plans/my-feature-plan.md
After each Codex invocation, extract session_id=xxx from the script output and save it as the session ID for the current task. In later Codex calls for the same task, pass --session <id> to reuse context so Codex remembers prior implementation and fix history instead of rereading the entire codebase every time.
Read the specified plan file and understand:
CLAUDE.mdIf the plan already contains a checklist (- [ ] / - [x]), use those items as execution units. If it does not define clear steps, split the work into reasonable batches, with no more than 5 file changes per batch.
Use the /codex skill and give Codex the following instruction:
Implement the code according to the plan in {plan-file-path}.
Current execution scope: {specific step or batch description}
Requirements:
- Follow the design in the plan exactly. Do not improvise beyond it.
- Obey the Code Quality Hard Limits defined in `CLAUDE.md`.
- Single file <= 800 lines, single function <= 50 lines, nesting <= 3 levels
- Run `pnpm build` after implementation to confirm compilation succeeds
- If the plan includes a checklist, mark completed steps as `[x]`
After implementation, list all changed files and provide a summary of each change.
After Codex finishes, I perform a code review. Important: I only read code and write reviews. I never directly modify source files.
pnpm build to confirm the compilation status.Append the review to reviews/{topic}-review.md (shared with plan-review):
---
## Code Review Round {N} — {YYYY-MM-DD}
**Scope**: {code scope covered in this review}
**Build Status**: PASS / FAIL
### Issues
#### Issue 1 ({severity}): {title}
**File**: {file-path:line}
{issue description}
**Fix**: {specific fix recommendation}
...
### Verdict: NEEDS_FIX / APPROVED
If Verdict: NEEDS_FIX, call /codex and have Codex fix the issues instead of editing them myself:
Read the latest Code Review round in {review-file-path}.
Check each issue one by one. Fix the valid issues, and explain why any disputed item is not actually a problem.
After making fixes, run `pnpm build` to confirm compilation succeeds.
List the issues that were fixed and the corresponding code changes.
If Verdict: APPROVED, skip to Step 6.
After Codex applies fixes, I review again, still without editing code directly:
Verdict: APPROVEDAfter each batch is completed, ask Codex to update the checklist in the plan file (- [ ] -> - [x]). If unfinished steps remain, go back to Step 2 for the next batch. Once all work is complete, move to the wrap-up.
Report the following to the user:
| Level | Meaning | Must Fix |
|---|---|---|
| Critical | Causes runtime failures or security vulnerabilities | Yes |
| High | Violates project conventions or has obvious design flaws | Yes |
| Medium | Code quality issue that should be improved | Recommended |
| Low | Style or preference issue | Optional |
| Suggestion | Optimization suggestion | Optional |
Verdict rules:
NEEDS_FIXAPPROVED with optional improvement notesplan-review: reviews/{topic}-review.md{topic} is the plan file name without .md## Round {N} for plan review and ## Code Review Round {N} for code reviewWeekly Installs
64
Repository
GitHub Stars
43
First Seen
12 days ago
Security Audits
Gen Agent Trust HubPassSocketPassSnykFail
Installed on
codex63
gemini-cli61
claude-code61
github-copilot59
opencode59
cline58
GitHub Actions 官方文档查询助手 - 精准解答 CI/CD 工作流问题
47,200 周安装
Google Gemini 文件搜索设置教程 - 完全托管RAG系统,支持100+格式,集成最佳实践
328 周安装
Go 代码风格核心原则与最佳实践 | Google/Uber 权威指南
332 周安装
Cloudflare Durable Objects 教程:使用 SQLite 和 WebSocket 构建有状态应用
332 周安装
Cloudflare D1数据库迁移指南 - Drizzle ORM工作流与问题解决方案
350 周安装
小红书内容转换器:一键将通用文章转为小红书爆款笔记格式 | AI写作助手
339 周安装
GTM市场进入策略指南:产品发布、新市场、重新定位与功能发布的完整蓝图
346 周安装