tech-review
Skill119 starsChanged 53 days agoMITView on GitHub
Reviews technical decisions before implementation. Detects outdated or suboptimal approaches by searching current best practices. Use when the user asks to implement something using a specific tool, library, or methodology.
5 other projects use a version of this.
Adopted by
Most recently taken up by
- Daeguk-Sun/dcNesstheir own version, first seen Nov 8, 2026
- azukiazusa1/skillstheir own version, first seen Oct 13, 2026
- k1ein-chen/Harness-Starterword for word, first seen Sep 24, 2026
- peterwangze/software-project-governancetheir own version, first seen Jun 9, 2026
- azukiazusa1/my-skill-repotheir own version, first seen Apr 19, 2026
Other versions5 projects use a version of this.
Word for word · 1
- k1ein-chen/Harness-Starter★ 119100% same
Edited · 0
None found.
Wrote their own version · 4
- azukiazusa1/skills★ 7same name
- peterwangze/software-project-governance★ 2same name
- Daeguk-Sun/dcNess★ 1same name
- azukiazusa1/my-skill-repo★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository8 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- handoverSkill
- harness-gcSkill
- harness-initSkill
- harness-modeSkill
- harness-startSkill
- verify-goalSkill
The file
Files
Tech Review
核心理念
用户因为信息差给出了过时或不优的技术方案时,AI 有责任在执行前发现并告知。
这不是"违抗用户指令"——不执行是违抗,发现更好的方案并告知是专业化。
触发条件
必须触发:用户指令中包含"用 [X] 做 [Y]"结构的技术实现请求。X 是一个工具/库/框架/方法论,Y 是一个目标。这种结构意味着用户已经做了技术选型,这个选型需要被验证。
不触发的典型场景:
- 纯业务逻辑("写个登录接口"——不指定具体工具)
- 纯 debug("修一下这个 bug")
- 纯配置变更("把端口改成 3000")
- 用户明确说"按我说的做,不用查"
- 你 100% 确定该方案仍是当前最佳做法
不确定要不要触发?宁查勿漏。多一次搜索远比让用户走弯路划算。
工作流
Step 1:提取技术决策
识别用户指令中所有"用 X 做 Y"性质的技术决策。可能不止一项。
Step 2:搜索验证
对提取出的每一项技术决策,执行 web search。搜索词不限以下示例,根据实际情况组织:
[X] alternative
[X] vs [已知替代方案]
[X] deprecated
[场景描述] best practice每次审查至少搜索一次。即使你认为自己的知识是准确的,也搜索确认——因为你不知道你的训练数据截止到什么时候。
Step 3:输出审查报告
使用以下固定格式输出结果:
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
🔍 技术方案审查
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
审查结果:[ ✅ 方案可行 / ⚠️ 有更好的选择 / ❌ 方案不推荐 ]
涉及的技术决策:
1. [选型 A] → [结论,附搜索来源概要]
2. [选型 B] → [结论,附搜索来源概要]
[审查结果为 ✅]
当前方案仍是最佳做法,按原计划执行。
[审查结果为 ⚠️ 或 ❌]
替代建议:
- 方案一:[名称] — [简要说明]
- 方案二:[名称] — [简要说明]
─────────────────────────────────────
请选择:A) 按原方案继续 B) 改用建议方案 C) 我自己调整
─────────────────────────────────────Step 4:等待选择
- A — 尊重用户意愿,按原方案执行,不再质疑
- B — 替换为建议方案,继续执行
- C — 用户修改需求后重新来过
- 用户不回复 — 不继续执行
说明
- 这个 Skill 的价值不在于"AI 替用户做决定",而在于消除信息差
- web search 会产生费用,但这是为了换取正确性
- 如果同一领域的同类决策已在本次对话中审查过,可以跳过重复搜索
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

