skills.rest
zachmayer
zachmayerCOMMUNITY
Explore all skills by this author →
Website𝕏@AI_Insight_SolsGitHub

claude

Offload prompts to a fresh Claude instance with full tool access.

2|Updated Oct 30, 2025
Search Tags:#automation#workflow#claude#reasoning#sub-agent#adversarial-review
One-click install
npx skills add https://github.com/zachmayer/skills --skill claude-zachmayer
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: claude
Source: https://github.com/zachmayer/skills/tree/main/.claude/skills/claude
Command: npx skills add https://github.com/zachmayer/skills --skill claude-zachmayer

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill solves the limitation of single-model context by offloading complex, time-consuming, or specialized tasks to a dedicated, fresh Claude sub-agent.

Core Features & Use Cases

  • Adversarial Self-Review: Use a second Claude instance to critique your code or prose for blind spots and logic errors.
  • Design and Taste Calls: Offload subjective design decisions or aesthetic critiques to a model optimized for global framing.
  • Background Processing: Execute long-running tasks in the background while you continue working on other priorities.

Quick Start

Ask the claude skill to perform an adversarial review of the current file to identify potential logic flaws or hidden biases.

Frequently Asked Questions about claude

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

FAQPage Schema
How do I run an adversarial review of my code to find logic flaws?▼

To run an adversarial review, offload your code or prose to a specialized Claude sub-agent. This fresh instance critiques the work to identify hidden biases, blind spots, and logic errors.

Can I execute long-running background tasks while continuing other work?▼

Yes, you can execute long-running background tasks by offloading them to a dedicated Claude sub-agent. This background processing mode allows you to continue working on other priorities simultaneously.

How does a sub-agent help with subjective design and taste critiques?▼

A sub-agent helps with design critiques by offloading subjective aesthetic decisions to a fresh Claude instance. This model evaluates your work using a global framing optimized for taste and design calls.

Do I need the Agent tool to orchestrate sub-agent tasks in Claude Code?▼

Yes, you need the Agent tool available within your Claude Code environment. The sub-agent orchestration requires this tool to facilitate the execution of specialized tasks with full tool access.

What is the best way to handle complex reasoning without losing single-model context?▼

The best way to handle complex reasoning is offloading it to a specialized Claude sub-agent. This solves single-model context limitations by spawning a fresh instance with full tool access for deep work.

Related Skills

Explore More Software Engineering Skills→
faisalanjumfaisalanjum
3

test-re-skill-tries-task

Test Task tool availability for spawning sub-agents in forked Skills.

Community
Basic
mkoziymkoziy

critic

Critique plans, prompts, or documents across five evaluation lenses.

Community
Advanced
ZpankzZpankz
7

research-router

Route research tasks to specialized sub-skills for structured analytical workflows.

Community
Advanced
kivo360kivo360
1

moai-cc-agents

Create and manage Claude Code sub-agents with structured agent files.

Community
Advanced