What problem does it solve? LLM context windows are limited, and large tool outputs, long-running plans, and multi-agent coordination quickly exhaust available tokens. This Skill provides patterns and utilities for using the filesystem as a persistent overflow layer so agents store, retrieve, and update context on demand instead of keeping everything in the prompt. ## Core Features & Use Cases - Tool Output Offloading: Automatically write large tool outputs (over ~2000 tokens) to scratch files and return compact references with summaries. - Plan Persistence: Save structured multi-step plans to disk so agents can re-read objectives and progress after context refreshes or summarization. - Sub-Agent Workspaces: Route sub-agent findings through per-agent directories instead of lossy message chains, with dynamic skill loading and terminal output persistence. - Use Case: A research agent runs a web search returning 8000 tokens. The Skill writes results to a scratch file, returns a 100-token reference, and the agent greps the file later for specific details. ## Quick Start Ask the agent to offload any tool output larger than 2000 tokens to a scratch file and return only a summary with the file path.
