context-engineering-collection

1# Agent Skills for Context Engineering
2 
3This collection provides structured guidance for building production-grade AI agent systems through effective context engineering.
4 
5## When to Activate
6 
7Activate these skills when:
8- Building new agent systems from scratch
9- Optimizing existing agent performance
10- Debugging context-related failures
11- Designing multi-agent architectures
12- Creating or evaluating tools for agents
13- Implementing memory and persistence layers
14- Designing autonomous research or evaluation harnesses
15 
16## Skill Map
17 
18### Foundational Context Engineering
19 
20**Understanding Context Fundamentals**
21Context is not just prompt text—it is the complete state available to the language model at inference time, including system instructions, tool definitions, retrieved documents, message history, and tool outputs. Effective context engineering means understanding what information truly matters for the task at hand and curating that information for maximum signal-to-noise ratio.
22 
23**Recognizing Context Degradation**
24Language models exhibit predictable degradation patterns as context grows: the "lost-in-middle" phenomenon where information in the center of context receives less attention; U-shaped attention curves that prioritize beginning and end; context poisoning when errors compound; and context distraction when irrelevant information overwhelms relevant content.
25 
26### Architectural Patterns
27 
28**Multi-Agent Coordination**
29…