Create, edit, review, and enhance AI prompts across multiple models.
npx skills add https://github.com/Fohlio/ai-tools --skill prompt-engineering-fohlio
Please help me install this Agent Skill. Skill: prompt-engineering Source: https://github.com/Fohlio/ai-tools/tree/main/skills/prompt-engineering Command: npx skills add https://github.com/Fohlio/ai-tools --skill prompt-engineering-fohlio
The Prompt Engineering Skill provides an evidence-based, model-aware framework for creating, editing, reviewing, and improving AI prompts across leading models. It standardizes the prompting process, reduces instruction drift, and accelerates onboarding for teams adopting multiple LLMs.
Ask for prompting help and specify the target model to tailor prompts.
High-intent search queries and answers about installing and using this skill.
Prompt optimization across models requires applying research-backed techniques and model-specific best practices to standardize instructions. This skill provides tailored guidance for Claude, GPT, Gemini, DeepSeek, and Qwen to reduce instruction drift.
Chain-of-thought prompting is a structured pattern that guides models through step-by-step reasoning before reaching a conclusion. Use it for complex tasks in coding, data analysis, or agent design where logical progression improves accuracy.
Yes, structured prompting patterns including ReAct, CTCO, CoT, and Few-shot are supported for building agent workflows and multi-step prompts. These patterns help standardize outputs and improve cross-model evaluation.
Cross-model evaluation involves using best-practice templates to audit prompts and adapt them to different model architectures. This skill provides evaluation workflows to ensure consistent performance across Claude, GPT, Gemini, DeepSeek, and Qwen.
Yes, it integrates with established prompt libraries including LangChain Hub, Anthropic Library, OpenAI Cookbook, and Google AI Studio. This allows you to leverage existing resources while applying evidence-based optimization techniques.
Formatting prompt outputs for data analysis is best achieved using structured JSON or XML schemas. This skill guides you in applying these structured output formats alongside techniques like self-consistency to ensure reliable data extraction.
Nickalus12Identify performance hotspots and regressions in Loom-style codebases.
bosens-ChinaBuild deep learning agents with LangChain/LangGraph middleware for planning and delegation.
Nuwanda04Analyze prompts, evaluate RAG systems, and design agentic workflows.
tomlupoOrchestrates a three-stage council of AI models for structured evaluation and synthesis.
