cavecrew

Delegate code analysis and editing tasks to specialized caveman-style subagents.

2|1|Updated Sep 25, 2025

Search Tags:#code analysis #context management #delegation #subagent #code editing

Ready to run

npx skills add https://github.com/istinataFTS/LiftLeagueLegends --skill cavecrew-istinatafts

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: cavecrew
Source: https://github.com/istinataFTS/LiftLeagueLegends/tree/main/fitness_tracker/.agents/skills/cavecrew
Command: npx skills add https://github.com/istinataFTS/LiftLeagueLegends --skill cavecrew-istinatafts

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

Helps users determine when to delegate specific tasks to caveman-style subagents, optimizing context and output size in code analysis and editing.

Core Features & Use Cases

Subagent Delegation: Offers a decision matrix for using subagents like cavecrew-investigator, cavecrew-builder, and cavecrew-reviewer.
Output Compression: Subagent output is compressed by 60% compared to vanilla prose, reducing context exhaustion.
Use Case: When analyzing a large codebase, cavecrew-investigator can quickly locate code paths, while cavecrew-builder can make precise edits.

Quick Start

Invoke 'cavecrew-investigator' to locate a symbol in the codebase.

Frequently Asked Questions about cavecrew

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

FAQPage Schema

How do I manage context exhaustion when analyzing a large codebase?▼

Context exhaustion during codebase analysis is managed by delegating tasks to specialized subagents. This approach compresses subagent output by 60% compared to vanilla prose, significantly reducing context window consumption while locating code paths.

When should I delegate code editing tasks to a subagent?▼

Code editing delegation to a subagent is recommended when you need precise edits in a large codebase and want to optimize context. A decision matrix triggers specialized agents like the builder for edits and the investigator for code path location.

How does subagent delegation work for code analysis?▼

Subagent delegation for code analysis works by triggering specialized agents based on a decision matrix. Agents like the investigator rapidly locate symbols and code paths, compressing output by 60% to prevent context exhaustion during large codebase analysis.

Can I use specific subagents for different code editing tasks?▼

Yes, specific subagents can be used for different code editing tasks. The system provides specialized agents such as an investigator to locate symbols, a builder to make precise edits, and a reviewer, ensuring efficient task specialization.

What is the best way to locate a specific symbol in a large codebase?▼

The best way to locate a specific symbol in a large codebase is to invoke a specialized investigator subagent. This delegates the code analysis task, compressing output by 60% to preserve context while rapidly finding the relevant code paths.

Why does delegating code analysis reduce context exhaustion?▼

Delegating code analysis reduces context exhaustion because subagent output is compressed by 60% compared to vanilla prose. By using a decision matrix to trigger specialized agents, context size is optimized during rapid code segment location and editing.