only-you

Identify frontier-worthy repository tasks beyond routine engineer or lower-tier model scope.

7|1|Updated May 30, 2026
One-click install
npx skills add https://github.com/yamadashy/skills --skill only-you
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: only-you
Source: https://github.com/yamadashy/skills/tree/main/skills/only-you
Command: npx skills add https://github.com/yamadashy/skills --skill only-you

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps identify and propose tasks that only the most capable models can handle, saving you time by filtering out tasks that other models or engineers can easily accomplish.

Core Features & Use Cases

  • Task Identification: Finds frontier-worthy tasks that the model one tier below or a routine engineer would struggle with.
  • Deliverable Pitch List: Provides a numbered list of tasks, each supported with evidence and reasons for its necessity.
  • Use Case: When you're unsure of what high-value tasks to tackle in your repository, the 'only-you' Skill can assist in prioritizing work that only a high-capable model should execute.

Quick Start

Use the only-you skill to analyze your repository and get a list of frontier-worthy tasks that can be performed at a high level.

Frequently Asked Questions about only-you

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

FAQPage Schema
How do I identify high-impact tasks in a software repository that require advanced model execution?

To identify high-impact tasks, you need to perform thorough repository analysis and compare potential tasks against those proposed by lower-tier models. This filters out routine work, leaving only frontier-worthy tasks that require a more capable model's execution.

What is a frontier-worthy task in software engineering?

A frontier-worthy task is a high-impact engineering challenge that goes beyond routine work or lower-tier model capabilities. It requires advanced reasoning and execution that only the most capable models can handle effectively.

How can I prioritize model-driven work to avoid wasting time on simple tasks?

You can prioritize model-driven work by comparing proposed repository tasks against what lower-tier subagents can achieve. This comparison filters out easily solvable items, ensuring you only execute tasks needing a higher-capability model.

Can I use repository analysis to find tasks that only a high-capable model should execute?

Yes, repository analysis can find tasks that only a high-capable model should execute. By evaluating the codebase and comparing tasks against lower-tier model proposals, you isolate complex frontier tasks.

What is the best way to filter out routine engineering work from complex repository tasks?

The best way to filter out routine engineering work is to compare your repository's task list against proposals generated by subagents. This highlights tasks that lower-tier models struggle with, isolating the high-impact work.