skills.rest
pnp
pnpOFFICIAL
Explore all skills by this author →
Website𝕏@m365pnpGitHub

prompt-context-engineer

Transforms rough prompts into structured, context-engineered prompts using component audits and templates.

864|298|Updated Apr 25, 2024
Search Tags:#prompt optimization#context engineering#system prompt#Templates#Prompt Engineering#Copilot#CoWork
One-click install
npx skills add https://github.com/pnp/copilot-prompts --skill prompt-context-engineer
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: prompt-context-engineer
Source: https://github.com/pnp/copilot-prompts/tree/main/samples/skills/prompt-context-engineer
Command: npx skills add https://github.com/pnp/copilot-prompts --skill prompt-context-engineer

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) and assets (resource) components.

What problem does it solve?

Rough or vague prompts produce inconsistent AI outputs because they lack the context components language models need, such as task framing, constraints, examples, and output format. This Skill diagnoses what a prompt is missing and rebuilds it into a structured, reusable prompt.

Core Features & Use Cases

  • Prompt Diagnosis: Audits a raw prompt against a nine-component checklist (task, role, background, audience, examples, constraints, output format, state, success criteria) and marks each as present, inferable, or unclear.
  • Guided Clarification: Infers sensible defaults for missing components and asks at most a few batched confirmation questions, with a speed mode that skips questions entirely.
  • Task-Type Templates: Builds the final prompt from skeletons in references/templates.md covering generation, transformation, extraction, classification, persona, and agentic system prompts.
  • Use Case: Paste a rough request like "summarize my meeting notes" and receive a copyable, structured prompt with fenced data slots, edge-case rules, and a checkable output format.

Quick Start

Ask Cowork to improve or engineer a pasted rough prompt, and it will return a structured, context-engineered prompt in a single copyable block.

Frequently Asked Questions about prompt-context-engineer

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I improve a rough prompt for an AI assistant?▼

Paste the rough prompt into a Cowork-enabled chat and ask to improve or engineer it. The skill audits missing context components, confirms a few inferred defaults, and returns a structured prompt in a single copyable code block.

What is context engineering in prompt writing?▼

Context engineering is filling the context window with exactly the information the task needs, such as task framing, background, examples, constraints, and output format, while excluding everything else. The skill applies this principle so every line of the prompt is load-bearing.

Can this skill write system prompts for agents and APIs?▼

Yes, it detects system prompt intent from cues like "system prompt", "API", or "agent" and uses the agentic template skeleton. Variable inputs are marked with {{PLACEHOLDERS}} and each slot is explained at the end of the prompt.

Does the prompt engineer skill answer the original prompt?▼

No, it only restructures the prompt itself and never executes the underlying task. If intent is ambiguous, it asks one short question to confirm whether you want the prompt rewritten or answered.

When should I not use prompt context engineering?▼

Skip it when the raw prompt is already complete and unambiguous, since the skill will only make marginal improvements. It is also not suited for general document writing or content generation unrelated to crafting prompts.

Related Skills

Explore More Content & Communication Skills→
InditexTechInditexTech
195

writing-fragments

Extract writing fragments from conversations and append them to a markdown document.

Official
Advanced
The-Entourage-AI-DevelopmentThe-Entourage-AI-Development

prompt-engineering-patterns

Design and deploy reusable prompt templates for LLM applications.

Official
Advanced
AdeitasunaAdeitasuna

eval-running

Run Loa evaluation suites to detect regressions and benchmark skill quality.

Community
Intermediate
EntityProcessEntityProcess
15

agentv-prompt-optimizer

Iteratively refine AI prompt files against AgentV evaluation datasets.

Official
Advanced