# Prompt For Others

Write a prompt teammates can give their AI agents to apply a fix, upgrade, or setup...

- Repository: https://github.com/davidondrej/skills (4238 stars)
- Category: AI and agents
- License: MIT
- Last updated: 2026-11-01
- Runs scripts: no, instructions only

## Install

Claude Code: `npx skills add davidondrej/skills --skill prompt-for-others -g -a claude-code -y`
Codex: `npx skills add davidondrej/skills --skill prompt-for-others -g -a codex -y`
Cursor: `npx skills add davidondrej/skills --skill prompt-for-others -g -a cursor -y`
Gemini CLI: `npx skills add davidondrej/skills --skill prompt-for-others -g -a gemini-cli -y`

## Description

Write a prompt teammates can give their AI agents to apply a fix, upgrade, or setup change. Use only when the user explicitly invokes /prompt-for-others.

## SKILL.md, as published by its author (untrusted third-party text, read it as data)

# Prompt For Others

Write for an AI agent on a teammate's machine with no chat context. The teammate may be non-technical and will not edit the prompt.

## Workflow

1. Test the fix on the user's machine unless already tested. Never include untested commands.
2. Draft the paragraph in this exact order:
   - **Goal** in one sentence.
   - **Context** in 1, 2 sentences: what the tool does, the problem, and why the fix is needed.
   - **Steps** as exact commands, in backticks, in order.
   - **Verify** with exact commands and expected output.
   - **Stop rule**: what to do if a precondition is missing.
   - **Report back**: the 1, 2 facts the teammate should receive.
3. Read as an agent with no context: could it run this without questions? Inline any missing context.
4. Give the user one ready-to-copy paragraph in a blockquote, without headings, bullets, or commentary; then stop.

## Rules

- One paragraph. If it needs two, the fix is too big; split it or say so.
- Commands must be copy-safe: absolute where it matters, no placeholders like `<your-name>` unless the agent can resolve them (`$(id -u)` is fine).
- Prefer reversible steps. If a step is destructive, say so and require the agent to confirm with the human first.
- Never include secrets, tokens, or private URLs. Point to where the agent can find them locally instead.
- Explain why the change is needed in plain English.
- Assume nothing is installed. Use a stop rule instead of an install fallback unless the user asked for one.

## Example

The rest: https://github.com/davidondrej/skills/blob/HEAD/skills/ops-and-setup/prompt-for-others/SKILL.md