mirror of
https://github.com/github/awesome-copilot.git
synced 2026-02-20 10:25:13 +00:00
c321e2a0a2
* feat: Add comprehensive AI Prompt Engineering & Safety Best Practices - Add 868-line instruction file covering prompt engineering fundamentals, safety, bias mitigation, security, and responsible AI usage - Add 527-line expert chat mode for prompt analysis and improvement with systematic evaluation frameworks - Include detailed templates, checklists, examples, and educational resources - Follow industry best practices from Microsoft, OpenAI, and Google AI principles - Implement comprehensive safety assessment frameworks and bias detection tools - Add advanced techniques: prompt chaining, templates, versioning, and testing frameworks - Provide extensive references to official guidelines, research papers, and community resources - Strictly comply with project guidelines (CONTRIBUTING.md, SECURITY.md, SUPPORT.md) Total: 1,395 lines of professional-grade prompt engineering guidance * feat: add comprehensive AI prompt engineering safety best practices instruction - Add 868-line comprehensive instruction file covering prompt engineering fundamentals - Include safety frameworks, bias mitigation, security best practices, and responsible AI usage - Provide detailed templates, checklists, and testing methodologies - Cover anti-patterns, prompt injection prevention, and ethical considerations - Add educational content on prompt patterns, validation techniques, and documentation - Include practical examples and real-world scenarios for developers - Rename file to avoid conflict with existing prompt-engineer chat mode - Update README with new instruction file This instruction provides developers with industry-standard best practices for creating safe, effective, and responsible AI prompts for Copilot and other LLMs. * fix: restore original prompt-engineer chat mode to match upstream - Restore the original prompt-engineer.chatmode.md from upstream/main - Remove overwritten comprehensive version to avoid duplication and conflict - Ensure only the official, simple prompt engineer chat mode is present All advanced prompt engineering best practices are now in the new instruction file only. * chore: regenerate README after restoring original prompt engineer chat mode - Re-run update-readme.js to ensure only the correct, original description for prompt-engineer.chatmode.md is present - Remove any duplicate or outdated descriptions from previous comprehensive version - README is now fully in sync with the restored chat mode and current repo state * feat: add comprehensive AI prompt engineering safety review prompt - Add 150+ line prompt file for AI prompt engineering safety review and improvement - Include systematic analysis framework covering safety, bias, security, and effectiveness - Provide structured output format with detailed assessment criteria - Include safety guidelines, testing recommendations, and educational insights - Complement the existing AI prompt engineering safety best practices instruction - Follow repository guidelines and best practices for prompt creation This prompt enables developers to systematically review and improve prompts for safety, bias mitigation, security, and responsible AI usage. * feat: expand AI prompt engineering safety review prompt with comprehensive analysis - Expand prompt from basic to comprehensive 200+ line analysis framework - Add advanced pattern analysis covering zero-shot, few-shot, chain-of-thought, role-based patterns - Include technical robustness assessment (input validation, error handling, scalability) - Add performance optimization metrics (token efficiency, response quality, consistency) - Enhance output format with detailed scoring and assessment criteria - Include bias mitigation strategies, security enhancements, and technical improvements - Provide extensive testing recommendations and educational insights - Follow repository guidelines for comprehensive, actionable prompt content This expanded prompt provides developers with a thorough, systematic approach to prompt safety review and improvement.