What problem does it solve? Squad agents load their full context history on every spawn, producing 34–74KB payloads where 82–96% is stale noise. This Skill defines a three-tier memory model that cuts context size by 20–55% per spawn while keeping relevant history and knowledge available on demand. ## Core Features & Use Cases - Three-Tier Memory Model: Hot tier (~2–4KB, always loaded), Cold tier (~8–12KB, on-demand cross-session summaries), and Wiki tier (durable structured knowledge like ADRs and conventions). - Spawn Template Pattern: Ready-to-use spawn prompt template that defaults to Hot-only loading with --include-cold and --include-wiki flags for selective expansion. - Scribe Integration: Automated tier promotion where the Scribe agent compresses Hot to Cold at session end and promotes Cold to Wiki after 30 days. - Use Case: When resuming interrupted debugging work, spawn an agent with Hot context plus the Cold summary from .squad/memory/cold/{agent-name}.md instead of replaying full transcripts. ## Quick Start Spawn an agent using the provided template with only the Hot memory section filled in, and add the Cold or Wiki sections only when the task requires history or domain knowledge.
