What problem does it solve? Subagent tool results are injected verbatim into the main conversation context, so verbose agent output quickly exhausts the context window during long coding sessions. This Skill provides a decision guide for delegating work to three caveman-compressed subagents whose output is roughly 60% smaller, letting the main context last longer. ## Core Features & Use Cases - Delegation Decision Guide: A lookup table tells you when to spawn cavecrew-investigator (locate code), cavecrew-builder (1-2 file edits), or cavecrew-reviewer (diff review) versus using vanilla agents or the main thread. - Structured Output Contracts: Each subagent returns a predictable, greppable format with file paths, line numbers, and severity totals, so results are easy to parse and chain. - Chaining Patterns: Supports locate-fix-verify pipelines, parallel investigator scouting, and single-shot edits when the target site is already known. - Use Case: During a long refactoring session, spawn cavecrew-investigator to find all call sites of a function, hand the top two sites to cavecrew-builder for surgical edits, then have cavecrew-reviewer audit the diff, all while consuming a fraction of the context a vanilla agent would. ## Quick Start Ask the assistant to use cavecrew to delegate locating a function definition to the investigator subagent instead of running a vanilla Explore agent.
