genshijin-crew

Select compressed subagents for code exploration, edits, and diff reviews.

304|10|Updated Apr 6, 2026
One-click install
npx skills add https://github.com/InterfaceX-co-jp/genshijin --skill genshijin-crew
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: genshijin-crew
Source: https://github.com/InterfaceX-co-jp/genshijin/tree/main/skills/genshijin-crew
Command: npx skills add https://github.com/InterfaceX-co-jp/genshijin --skill genshijin-crew

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps you decide when to delegate work to specific compressed subagents so you can save main-context tokens while still performing accurate codebase exploration, small edits, and diff reviews.

Core Features & Use Cases

  • Delegation decision guide: Chooses between investigator, builder, and reviewer based on whether locations are known, how many files are involved, and whether you need structured findings or surgical changes.
  • Structured output contracts: Enforces predictable, grep-friendly formats for investigator results, builder edits, and reviewer issue reporting to reduce wasted context.
  • Token-efficiency rationale: Targets reduced tool-result verbosity (about 60% shrink per delegation) to sustain long sessions and avoid context exhaustion.

Quick Start

Use genshijin-crew when you want the AI to delegate investigation/edit/review steps to compressed subagents and return structured findings instead of long prose.

Frequently Asked Questions about genshijin-crew

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

FAQPage Schema
How do I reduce token usage during codebase exploration?▼

To reduce token usage during codebase exploration, delegate search tasks to compressed subagents that return structured, grep-friendly findings instead of long prose, shrinking tool results by about 60%.

When should I delegate code edits to a subagent instead of the main context?▼

Delegate code edits to a builder subagent when target file locations are already identified, enabling surgical changes that preserve decision quality while preventing wasted main-context tokens.

What is the best way to orchestrate code review without exhausting context?▼

Orchestrate code review without exhausting context by routing diff inspections to a reviewer subagent, which enforces structured issue reporting formats to minimize verbosity and sustain long sessions.

Does workflow orchestration with subagents require known file locations?▼

Workflow orchestration with subagents requires known file locations specifically for the builder role; an investigator subagent handles code exploration when locations are unknown, guided by delegation guardrails.

Why does context compression matter for long coding sessions?▼

Context compression matters for long coding sessions because formatting contracts reduce tool-result verbosity by approximately 60%, preventing context exhaustion and sustaining decision quality.

Can I use structured outputs for diff review and code search?▼

You can use structured outputs for diff review and code search by enforcing predictable, grep-friendly formats through subagent delegation, ensuring efficient main-context return for investigators and reviewers.