How does it handle behavioral differences in Opus 4.5?

It provides optional prompt adjustments to mitigate issues like tool overtriggering, over-engineering, and 'thinking' sensitivity when extended thinking is disabled.

Which platforms are supported for model string updates?

The skill supports model string updates for the native Anthropic API, AWS Bedrock, Google Vertex AI, and Azure AI Foundry.

Does this skill migrate Claude Haiku models?

No, this skill is specifically designed for migrations to Opus 4.5 and explicitly excludes Haiku 4.5 migration tasks.

What is the 'effort' parameter mentioned in the migration?

Opus 4.5 supports an 'effort' parameter to control reasoning depth; this skill defaults to 'high' for migration tasks to ensure maximum accuracy.

Claude Opus 4.5 Migration Helper

Name: Claude Opus 4.5 Migration Helper
Author: Microck

byMicrock

•

Infraestructura en la Nube

Automates the transition of codebases and prompts from legacy Claude models to the Claude 3 Opus 4.5 architecture.

Acerca de

This skill provides a comprehensive workflow for upgrading AI-integrated applications to Claude Opus 4.5. It automatically identifies and replaces legacy model strings across multiple platforms—including the Anthropic API, AWS Bedrock, Google Vertex AI, and Azure AI Foundry—while removing deprecated beta headers. Beyond simple updates, the skill offers specialized logic to refine system prompts, addressing Opus 4.5's unique behavioral characteristics such as tool overtriggering, over-engineering tendencies, and sensitivity to 'thinking' parameters, ensuring a stable and optimized migration.

Características Principales

Targeted snippets for code exploration and frontend design optimization
Automatic configuration of the 'high' effort parameter for complex tasks
Intelligent prompt softening to prevent tool overtriggering
Removal of unsupported context-1m-2025-08-07 beta headers
Automated model string replacement for 1P, Bedrock, Vertex, and Azure platforms
89 GitHub stars

Casos de Uso

Standardizing model identifiers across a multi-cloud (AWS/GCP) AI architecture
Fine-tuning prompt behavior after experiencing over-engineering issues with newer models
Upgrading an enterprise application from Claude 3.5 Sonnet to Opus 4.5

Acerca de

Características Principales

Targeted snippets for code exploration and frontend design optimization
Automatic configuration of the 'high' effort parameter for complex tasks
Intelligent prompt softening to prevent tool overtriggering
Removal of unsupported context-1m-2025-08-07 beta headers
Automated model string replacement for 1P, Bedrock, Vertex, and Azure platforms
89 GitHub stars

Casos de Uso

Standardizing model identifiers across a multi-cloud (AWS/GCP) AI architecture
Fine-tuning prompt behavior after experiencing over-engineering issues with newer models
Upgrading an enterprise application from Claude 3.5 Sonnet to Opus 4.5