Context Engineering Advisor

Diagnose context stuffing and implement context engineering for AI interactions.

1|Updated Mar 11, 2026
One-click install
npx skills add https://github.com/zhengxuyu/default-talents --skill context-engineering-advisor-zhengxuyu
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: Context Engineering Advisor
Source: https://github.com/zhengxuyu/default-talents/tree/main/product-manager-skills/skills/context-engineering-advisor
Command: npx skills add https://github.com/zhengxuyu/default-talents --skill context-engineering-advisor-zhengxuyu

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill helps product managers distinguish between effective "context engineering" and inefficient "context stuffing" when working with AI, leading to more focused, accurate, and cost-effective AI interactions.

Core Features & Use Cases

  • Diagnose Context Stuffing: Identify symptoms of jamming too much irrelevant information into AI prompts.
  • Context Engineering Principles: Learn to structure information for AI attention, manage memory, and implement efficient cycles.
  • Use Case: A PM is pasting entire PRDs and user interview transcripts into an AI, getting vague answers. This Skill guides them to identify only the essential context needed for a specific decision, drastically improving AI output quality and reducing token costs.

Quick Start

Use the Context Engineering Advisor skill to diagnose your current AI context practices and receive tailored recommendations.

Frequently Asked Questions about Context Engineering Advisor

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

FAQPage Schema
What is context engineering versus context stuffing in AI prompt interactions?

Context engineering structurally organizes information for AI attention to prevent reasoning noise, whereas context stuffing jams irrelevant data into prompts. It treats AI attention as a scarce resource to improve output quality.

How do I stop getting vague AI answers when pasting entire PRDs and user interview transcripts?

Diagnose your prompt context boundaries and apply the Research→Plan→Reset→Implement cycle. Identifying only essential decision-specific context eliminates reasoning noise from context stuffing and reduces token usage.

How does the Research Plan Reset Implement cycle improve LLM efficiency?

The Research→Plan→Reset→Implement cycle manages memory architecture by segmenting AI interactions. It isolates research and planning phases before resetting context, ensuring focused implementation and efficient token consumption.

Does context engineering require prerequisite knowledge of information architecture design?

Yes, context engineering requires understanding information architecture design to structure inputs for AI attention. Product managers must grasp how to diagnose context boundaries and manage memory architecture effectively.

What are the symptoms of context rot and how do I identify context stuffing in my prompts?

Symptoms of context rot include reasoning noise, vague outputs, and inefficient token usage. You identify context stuffing by diagnosing whether your prompt contains irrelevant information that overwhelms the AI's attention boundaries.