cavecrew

Guide the use of caveman-style subagents for code investigation, editing, and review.

Updated May 17, 2026
One-click install
npx skills add https://github.com/galeep/plugin-place --skill cavecrew-galeep
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: cavecrew
Source: https://github.com/galeep/plugin-place/tree/main/plugins/caveman/skills/cavecrew
Command: npx skills add https://github.com/galeep/plugin-place --skill cavecrew-galeep

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill helps streamline decision-making processes by guiding when to use caveman-style subagents for tasks such as code investigation, editing, and review, compared to vanilla equivalents.

Core Features & Use Cases

  • Task-Specific Subagents: Offers three specialized subagents for specific code-related tasks, each optimized for compressed output.
  • Context Management: Ensures main context preservation by using compressed agent output, reducing the risk of context exhaustion.
  • Use Case: When working with large codebases, you can use this Skill to delegate complex investigations to cavecrew-investigator, surgical edits to cavecrew-builder, and code reviews to cavecrew-reviewer.

Quick Start

To invoke cavecrew, mention 'delegate to subagent' or 'use cavecrew' during a conversation.

Frequently Asked Questions about cavecrew

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I preserve context when investigating and editing large codebases?

To preserve context in large codebases, you can delegate complex code investigation, surgical edits, and code review tasks to specialized subagents that return compressed output, reducing the risk of context exhaustion and lowering token usage.

What is a caveman-style subagent for code review and editing?

A caveman-style subagent is a specialized agent optimized for compressed output, designed to handle specific code-related tasks like investigation, editing, and review while preserving the main context of the session.

How do I delegate code tasks to subagents to reduce token usage?

You can delegate code tasks to subagents by invoking the skill during a conversation, mentioning 'delegate to subagent' or 'use cavecrew' to route investigations, edits, and reviews to specialized agents for reduced token consumption.

Can I use specialized subagents for surgical code edits in large codebases?

Yes, you can use specialized subagents for surgical code edits in large codebases by delegating the editing tasks to a dedicated builder agent, ensuring main context preservation and optimized efficiency.

When should I use subagents instead of vanilla agents for code investigation?

You should use subagents instead of vanilla equivalents when working with large codebases that require context preservation across sessions, as specialized subagents provide compressed output to prevent context exhaustion.

Does managing context across sessions require specialized code review subagents?

Managing context across sessions benefits from specialized code review subagents, which handle reviews and return compressed output to maintain main context preservation and optimize overall decision-making efficiency.