What problem does it solve? AI coding agents produce poor output when they lack the right context, leading to hallucinations, wrong patterns, and stale assumptions. This Skill provides a structured strategy for feeding agents the correct rules, skills, MCP data, and project files at the right time. ## Core Features & Use Cases - Context Hierarchy Management: Organizes context into always-active rules, auto-loaded background skills, and on-demand active skills invoked via slash commands. - MCP Integration Strategy: Uses DaaS and Buildpad MCP servers to query live schema, permissions, and component data instead of relying on stale assumptions. - Quality Drop Recovery: Provides a diagnostic decision tree for recovering from degraded agent output by loading the right skill or refreshing context. - Use Case: When starting a new coding session on a Next.js and Supabase project, follow the session-start checklist to read environment config, check project phase progress, and load relevant background skills before writing code. ## Quick Start Ask the agent to load the context-engineering skill and apply its session-start checklist before beginning a new coding task.
