external-model-selection

The 'external-model-selection' skill is designed to help developers and teams choose the best external AI models for specific tasks such as code analysis, bug investigation, and architectural decisions. By consulting multiple large language models (LLMs), this skill enables users to select optimal models based on empirically validated performance data. Whether you need to compare different AI perspectives for debugging issues, make architectural decisions, or investigate complex problems, this skill helps streamline the selection process by providing rankings and recommendations that have been tested in real-world scenarios.

This skill offers a variety of powerful features, including fast and cost-effective model selection, detailed scoring of models based on their performance in various use cases, and proven strategies for consultation. The skill includes top-performing models like MiniMax M2, Grok Code Fast, and GPT-5.1 Codex, each recommended for specific tasks such as bug diagnosis, code debugging, and architectural planning. The system ranks these models on factors like speed, cost, and accuracy, with real-world examples demonstrating their success in resolving complex technical challenges. Additionally, the inclusion of free models such as Sherlock Think Alpha and powerful new models like Gemini 3 Pro Preview ensures that users have access to a wide range of tools for every type of analysis.

This skill is perfect for developers, software engineers, and technical teams working on complex codebases, architectural redesigns, or bug investigations. It is especially valuable for those who need quick, reliable AI insights or are working with large-scale projects that require comprehensive analysis. Whether it's debugging a production issue, validating test coverage, or planning long-term software development, the external-model-selection skill provides the tools necessary to make informed decisions with high efficiency and accuracy.