whiteboard-explain
Use when the user asks to explain or teach a technical concept. Replies in ASD-STE100 Simplified Technical English with a simple diagram instead of a wall of jargon.
How do I install this agent skill?
npx skills add https://github.com/lwlee2608/agent-skills --skill whiteboard-explainIs this agent skill safe to install?
- Gen Agent Trust Hubpass
The skill is safe. It contains only natural language formatting instructions and examples for explaining technical concepts using Simplified Technical English and text-based diagrams. It does not execute code, handle credentials, or communicate over a network.
- Socketpass
No alerts
- Snykpass
Risk: LOW · No issues
What does this agent skill do?
Whiteboard Explain
Explain the concept as a good engineer explains it at a whiteboard. The goal is only this: the user understands.
Write the reply in ASD-STE100 Simplified Technical English.
Language rules
- Write short sentences. Use a maximum of 20 words in a sentence.
- Write short paragraphs. Use a maximum of 6 sentences in a paragraph.
- Use the active voice. Write "the balancer sends the request". Do not write "the request is sent".
- Give one instruction in one sentence.
- Use one word for one meaning. Do not change the word for the same thing later in the reply.
- Use the simple tenses: simple present, simple past, and simple future.
- Do not use the -ing form as a verb or as a noun. Write "when the server starts". Do not write "on starting the server".
- Use articles. Write "the server". Do not write "server".
- Do not put more than three nouns together. Write "the table of the message queue". Do not write "the message queue table".
- Do not use slang, idioms, or unusual words.
- Make a vertical list when the text has more than three related items.
- Keep the technical names and the technical verbs of the subject. Use each one the same way each time.
Whiteboard rules
- Make a small diagram when the diagram is more clear than the words.
- Explain one concept in one reply. Do not write an introduction or a summary.
Worked example
User: "Explain how a load balancer works."
Good:
The load balancer sends each request to one server.
┌──────────┐
│ Client │
└────┬─────┘
▼
┌───────────┐
│ Load │ ← selects one server for each request
│ Balancer │
└─────┬─────┘
┌──────┼──────┐
▼ ▼ ▼
┌───┐ ┌───┐ ┌───┐
│S1 │ │S2 │ │S3 │
└───┘ └───┘ └───┘
- The client sees one address. The client does not know which server sent the answer.
- If server S2 stops, the balancer does not send more requests to S2.
- If the traffic increases, add server S4 and server S5. The client does not change.
Bad — a wall of jargon, and no diagram:
A load balancer is a network device or software component that distributes incoming traffic across multiple backend servers using algorithms such as round-robin, least-connections, IP-hash…
Bad — the passive voice, a long sentence, and the -ing form:
Requests are distributed by the load balancer to the available servers, with unhealthy servers being removed from the pool automatically after failing a configurable number of consecutive health checks.
How can the creator link this skill?
Add the canonical catalog link to the repository README so users can inspect current installs and available audits. The publishing guide covers the complete discovery path.
<a href="https://skillzs.dev/skills/lwlee2608/agent-skills/whiteboard-explain">View whiteboard-explain on skillZs</a>