Identify when to delegate debugging tasks to compressed subagent outputs.
npx skills add https://github.com/jchildree/The-QUESTION --skill cavecrew-jchildree
Please help me install this Agent Skill. Skill: cavecrew Source: https://github.com/jchildree/The-QUESTION/tree/main/.agents/skills/cavecrew Command: npx skills add https://github.com/jchildree/The-QUESTION --skill cavecrew-jchildree
The Cavecrew skill unit solves the problem of information overload during complex debugging sessions by suggesting the use of compressed, subagent-driven output when it is appropriate.
cavecrew-investigator, cavecrew-builder, and cavecrew-reviewer.Explore or Code Reviewer, use Cavecrew to get compact findings that can be acted upon more efficiently.To determine whether to use cavecrew, consider phrases like "delegate to subagent" or "use cavecrew".
High-intent search queries and answers about installing and using this skill.
To preserve main thread context during long debugging sessions, delegate information retrieval and investigation tasks to compressed subagent outputs. This prevents information overload by returning concise findings instead of lengthy responses.
Compressed subagent output is a concise format returned by delegated subagents during code investigation. Use it during prolonged bug investigations when lengthy responses from standard tools would overwhelm your available context window.
Delegate tasks to subagents by invoking specialized roles like investigator, builder, or reviewer. These subagents process code review requests and return compact findings that can be acted upon more efficiently than full-length outputs.
Subagent-driven output is suitable for large-scale bug investigations requiring long-term context conservation. It selectively delegates information retrieval to maintain efficiency, making it ideal for environments needing concise task delegation over extended periods.
Compressed subagent output is limited to scenarios requiring concise information retrieval and task delegation. It should not be used when full-length, detailed responses from standard exploration or code review tools are necessary for the investigation.
Context conservation matters when debugging complex software because lengthy automated responses cause information overload. By selectively using compressed subagent output, you maintain available context and improve efficiency throughout prolonged investigation sessions.
shirulotDelegate coding tasks to specialized subagents with compressed output contracts.
achmfParse Sentry bot comments and patch code issues in GitHub pull requests.
phucle297Guide subagent delegation for code investigation, file editing, and code review.
wubq511Generate complete LLM outputs without truncation or placeholder shortcuts.