Menu
Craft effective prompts that get the best results from AI assistants including Copilot itself.
The review-first prompt is selected by default. You can switch to a direct command or download a local copy.
Read SKILL.md and any companion files shown by SkillsMP before deciding whether to install.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
A direct command skips the review prompt. Inspect the source before running it.
npx skills add https://github.com/fabioc-aloha/Alex_Plug_In --skill prompt-engineeringThe command stays on one line. Scroll horizontally to inspect it before copying.
Prefer a local copy? Download the files currently available to SkillsMP.
Showing SKILL.md
Create and maintain ASCII visual dashboards for project tracking with parallel lane progress bars
Store and manage voice samples for TTS cloning — portable, version-controlled audio references
Clear documentation through visual excellence
Based on SOC occupation classification
| name | Prompt Engineering |
| description | Craft effective prompts that get the best results from AI assistants including Copilot itself. |
Write better prompts to get better results. The quality of AI output depends directly on input quality. This skill teaches how to structure requests so that AI assistants produce accurate, relevant, and useful responses on the first attempt.
| Element | Description | Example |
|---|---|---|
| Context | Background information the AI needs | "I'm a project manager writing a status report for senior leadership" |
| Role | Who the AI should act as | "Act as a financial analyst" |
| Action | What specifically to do | "Summarize the key risks from this document" |
| Format | How to structure the output | "Use a table with columns: Risk, Likelihood, Impact, Mitigation" |
| Tone | Voice and style | "Professional but concise, suitable for C-suite audience" |
"Create a [document type] for [audience] that covers [topics]. Format as [structure]. Keep it [length/tone]."
Example: "Create a project status update for the steering committee that covers timeline, budget, and risks. Format as a one-page summary with RAG ratings. Keep it concise and action-oriented."
"Here is an example of what I want: [paste example]. Now create a similar one for [new context]."
"Walk me through [process] step by step. For each step, explain what to do and what to watch out for."
"Here is my draft: [paste text]. Identify three weaknesses and rewrite to fix them."
"Analyze [topic] from three perspectives: [role 1], [role 2], [role 3]. For each, list top 3 concerns."
| Problem | Cause | Fix |
|---|---|---|
| Too vague | "Write something about our project" | Specify: what document, for whom, covering what |
| Too broad | "Tell me everything about risk management" | Narrow: "List the top 5 risks for a cloud migration project" |
| No format specified | AI chooses its own structure | State: "as a table" or "as bullet points" or "max 200 words" |
| Missing context | AI fills gaps with assumptions | Provide: audience, purpose, constraints |
| One-shot expectation | Expecting perfection on first try | Iterate: prompt, review, refine |
| Technique | When to Use | How |
|---|---|---|
| Chain of thought | Complex analysis | "Think through this step by step before giving your answer" |
| Constraints | Controlling scope | "In under 200 words" / "Using only data from Q3" |
| Negative instructions | Preventing unwanted output | "Do not include generic advice. Only cite specific data." |
| Output seeding | Controlling structure | Start the output yourself: "The three key findings are: 1." |
| Persona stacking | Multiple viewpoints | "First as a CFO, then as a customer. Compare their concerns." |
