transformer

Rewrite existing SKILL.md instructions for frontier reasoning model calibration.

10|5|Updated Mar 3, 2026
One-click install
npx skills add https://github.com/JHostalek/dotclaude --skill transformer-jhostalek
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: transformer
Source: https://github.com/JHostalek/dotclaude/tree/main/skills/transformer
Command: npx skills add https://github.com/JHostalek/dotclaude --skill transformer-jhostalek

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill improves existing agent skills that are overly verbose, over-scaffolded, poorly calibrated, or misaligned with frontier reasoning models, while preserving the instructions that carry essential behavior.

Core Features & Use Cases

  • Full Skill Rewrites: Reframe or replace an existing skill when incremental edits would leave its underlying structure incoherent.
  • Frontier-Model Calibration: Reduce redundant verification, delegation pressure, narration, scope expansion, and excessive output while adding targeted guidance where newer models need it.
  • Context and Structure Review: Identify contradictions, repetition, misplaced tool guidance, weak interfaces, and opportunities for progressive disclosure.
  • Use Case: Apply this Skill to an audit workflow that has accumulated generic checklists and verifier-agent steps, transforming it into a concise, opinionated process with clear coverage requirements and a stronger output contract.

Quick Start

Use the transformer skill to rewrite the skill at the provided SKILL.md path for better frontier-model performance.

Frequently Asked Questions about transformer

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

FAQPage Schema
How do I rewrite existing agent skills for better frontier model judgment?

To rewrite existing agent skills for better frontier model judgment, read the target SKILL.md completely, preserve load-bearing interfaces, consult the bundled Opus 5 reference, and rewrite the file in place to improve calibration and operational coherence.

What is progressive disclosure in context engineering for agent workflows?

Progressive disclosure in context engineering is a structural technique that layers context cleanly, removing contradictions and repetition so agent workflows receive only the guidance needed at each stage.

How do I reduce excessive output and redundant verification in Claude Code skills?

Reducing excessive output and redundant verification in Claude Code skills requires a full skill rewrite that strips over-scaffolding, delegation pressure, and scope expansion while adding targeted guidance for frontier reasoning models.

When should I do a full skill rewrite instead of incremental edits?

A full skill rewrite is needed when incremental edits would leave the underlying structure incoherent, such as when an audit workflow has accumulated generic checklists and verifier-agent steps that require a concise, opinionated process.

Does the transformer skill create new prompt artifacts from scratch?

No, the transformer skill does not create new prompt artifacts. It applies exclusively to existing skill instructions that need structural redesign, context-layer cleanup, or adaptation for frontier reasoning models like Opus 5.