3w
Skill2 starsChanged 2 months agoMITView on GitHub
Break a possible adjustment or change down into Where, What, Why — where it applies, what actually changes, why it matters — plus what to do about it.
In this repository10 other files it gives its agents.
- catchupSkill
- clear-viewSkill
- decisionSkill
- first-principleSkill
- impactSkill
- one-lineSkill
- scopeSkill
- tldrSkill
- tradeoffSkill
- visSkill
The file
We split this skill into the rules it gives an agent
5 rules. About 0 are common, 5 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈5
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 26
If the adjustment touches multiple places, list each Where/What pair, but keep one shared Why unless they differ.
- Distinctive: line 63
Don't write "Where: the auth system" — that's a restatement of the request.
- Distinctive: line 66
Don't write "Why: it's a security best practice." Say what the current behavior actually costs.
- Distinctive: line 69
Don't let the bottom line echo Why ("Why: limits exposure" → "Bottom line: worth doing for the security gain").
- Distinctive: line 73
If no request was supplied, ask: "3W of what adjustment?"
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
3W
Goal: engineer sees exactly where a change lands, what it does, why it's worth doing, and what to do next — in one glance.
Rules
- Respond in the same language the user wrote in. If they write in Chinese, reply in Chinese. Technical terms (JWT, Redis, API…) stay in English.
- Max 10 lines. Hard limit.
- Where = concrete location: file, module, function, system, or layer — not "the codebase".
- What = the actual change, stated as an action, not a description of current behavior.
- Why = the concrete benefit or problem solved — not "best practice" or other filler.
- Bottom line = what to actually do now. It answers a different question than Why: Why is "is this worth doing", bottom line is "so what happens next". Use it for the sequencing trap, the prerequisite, or the reason to hold off — never to restate Why.
- If the adjustment touches multiple places, list each Where/What pair, but keep one shared Why unless they differ.
- No emoji.
Output format
**Where**
[file/module/function/system]
**What**
[the concrete change, one action]
**Why**
[the concrete reason — problem solved or benefit gained]
**Bottom line:** [what to do now — the sequencing trap, prerequisite, or hold-off reason]Example
/3w 把 access token 的有效期從 24 小時縮到 15 分鐘
**Where**
`auth/session.ts`, token refresh middleware
**What**
Shorten access token expiry from 24h to 15min, add silent refresh
**Why**
Stolen tokens currently stay valid for a full day — shrinking the window limits exposure
**Bottom line:** do it, but ship silent refresh in the same deploy or every user gets logged out hourlyAnti-examples
Don't write "Where: the auth system" — that's a restatement of the request. Name the file or function someone would open.
Don't write "Why: it's a security best practice." Say what the current behavior actually costs.
Don't let the bottom line echo Why ("Why: limits exposure" → "Bottom line: worth doing for the security gain"). If you have nothing new to add, the bottom line is the prerequisite or the order of operations.
If no request was supplied, ask: "3W of what adjustment?"
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.

