What problem does it solve? Subagent tool results are injected verbatim into the main conversation context, so verbose agent output quickly exhausts the context budget during long sessions. This Skill provides a decision guide for when to delegate work to three compressed-output subagents instead of working inline or using default agents. ## Core Features & Use Cases - Decision Matrix: Maps task types (locate code, surgical edit, diff review) to the right subagent: cavecrew-investigator, cavecrew-builder, or cavecrew-reviewer. - Output Contracts: Defines compact, greppable return formats (path:line findings, severity-tagged review findings) so results consume fewer main-context tokens. - Chaining Patterns: Documents workflows like locate → fix → verify and parallel scouting with multiple investigators. - Use Case: During a long debugging session, spawn cavecrew-investigator to find all callers of a function, hand the best site to cavecrew-builder for a two-file fix, then have cavecrew-reviewer audit the diff — all while keeping main context lean. ## Quick Start Ask the assistant to delegate code investigation to a cavecrew subagent instead of exploring inline when you want compressed results.
