Add model-recommendation prompt to suggest optimal Copilot models for chatmodes/prompts

[difegam]

Introduces a detailed prompt for recommending optimal AI models based on .chatmode.md or .prompt.md file analysis. The prompt outlines a workflow for task evaluation, model selection criteria, and cost-efficiency considerations. It includes support for subscription tiers, deprecated model migration, and advanced features like Context7 integration.

This addition enhances the ability to automate and optimize AI model recommendations, improving alignment with task complexity, user priorities, and available tools.

Pull Request Checklist

• I have read and followed the CONTRIBUTING.md guidelines.
• My contribution adds a new instruction, prompt, or chat mode file in the correct directory.
• The file follows the required naming convention.
• The content is clearly structured and follows the example format.
• I have tested my instructions, prompt, or chat mode with GitHub Copilot.
• I have run npm start and verified that README.md is up to date.

---

Description

This PR adds prompts/model-recommendation.prompt.md, a reusable agent prompt that analyzes a .chatmode.md or .prompt.md file and recommends the best GitHub Copilot model(s) based on task complexity, reasoning depth, context needs, tool usage, and subscription tier. It produces a structured “AI Model Recommendation Report” with a primary pick, alternatives, cost implications, auto-selection guidance, deprecation warnings, and frontmatter update suggestions. When needed, it can verify current capabilities and deprecations via Context7.

Context & Usage Notes

• Location: prompts/model-recommendation.prompt.md
• Purpose: Selects the most suitable model (quality/speed/cost) for a given chat mode or prompt by analyzing frontmatter, tools, and instructions.
• Inputs:
  • Required: filePath → path to the .chatmode.md or .prompt.md to analyze
  • Optional: subscriptionTier → Free | Pro | Pro+ (default: Pro)
  • Optional: priorityFactor → Speed | Cost | Quality | Balanced (default: Balanced)
• Tools declared: search/codebase, fetch, context7/*
  • Uses cost‑effective 0x models for read‑only tasks; elevates to premium models for complex reasoning or long‑context needs.
  • Can validate model features, multipliers, and deprecations with Context7 when uncertainty exists.
• Output: A structured report including:
  • File summary and task analysis (complexity, reasoning depth, context size, code‑gen needs)
  • Primary model + 1–2 alternatives with trade‑offs
  • Cost impact by subscription tier
  • Auto‑selection guidance (when to rely on Auto vs. manually selecting a model)
  • Deprecation checks and migration paths
  • Frontmatter update snippet ready to paste

---

Type of Contribution

• New instruction file.
• New prompt file.
• New chat mode file.
• New collection file.
• Update to existing instruction, prompt, chat mode, or collection.
• Other (please specify):

---

Additional Notes

• Frontmatter sets model: Auto (copilot) by default; the report explains when to keep Auto vs. manually choosing a specific model.
  • Includes current deprecation guidance (e.g., o3, o4‑mini, Gemini 2.0 Flash) and recommended replacements.
  • Before submitting, run the repo’s README update task so the new prompt appears in README.md as required by CONTRIBUTING.

---

By submitting this pull request, I confirm that my contribution abides by the Code of Conduct and will be licensed under the MIT License.

[Copilot]

Pull Request Overview

This PR adds a new AI model recommendation prompt that helps users analyze .chatmode.md or .prompt.md files to determine the optimal AI model for their specific use case. The prompt provides comprehensive guidance on model selection based on task complexity, required capabilities, cost-efficiency, and subscription tier constraints.

Key Changes

• Introduces a comprehensive model recommendation system with detailed decision trees and capability matrices for GitHub Copilot's available AI models
• Adds structured workflow for analyzing chat modes and prompts, including task categorization, model evaluation, and integration recommendations
• Includes up-to-date model information (as of October 2025) with deprecation notices and subscription tier considerations

Reviewed Changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated 1 comment.

File	Description
prompts/model-recommendation.prompt.md	New 677-line prompt file providing AI model recommendation capabilities with extensive documentation on model selection criteria, decision trees, and implementation guidance
docs/README.prompts.md	Updated prompts index to include entry for the new model recommendation prompt with installation badges and description

[Copilot]

The model field value should not include parenthetical explanations. Based on the coding guidelines and other prompt files, this should be either a specific model name (e.g., 'Claude Sonnet 4') or 'Auto' without additional text. The value 'Auto (copilot)' appears to be mixing the model selection with an explanation.

[difegam]

@copilot open a new pull request to apply changes based on the comments in this thread

[aaronpowell]

@difegam just so I'm understanding the intent - the prompt is designed to give the user an insight into why a certain model would (potentially) be selected if they use Auto, or a model that they should specify (and why) if they want to define a model in the model property of the front matter.

[difegam]

Hi @aaronpowell, the prompt will analyze the chat mode or prompt file to identify which of the available models might perform best based on criteria defined in the model comparison documentation at the GitHub Copilot reference.

• https://docs.github.com/en/copilot/reference/ai-models/model-comparison
• https://gist.github.com/difegam/6a52b57dbbbf38708017c6bc730e0ed1

In other words, if you want to specify a model in the model property or compare the results generated by different models, you can use this prompt to receive suggestions for a model that, based on the prompt/chat mode task, could offer the best performance and the reasons behind it.