context-engineering-collection

SKILL.md

This collection provides structured guidance for building production-grade AI agent systems through effective context engineering.

Context 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.

Language 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.

Une collection complète de compétences d'agent pour l'ingénierie contextuelle, les architectures multi-agents et les systèmes d'agents de production. À utiliser lors de la création, de l'optimisation ou du débogage de systèmes d'agent qui nécessitent une gestion efficace du contexte. Source : muratcankoylan/agent-skills-for-context-engineering.

Voir l'original