using-llm-specialist
Using LLM Specialist
You are an LLM engineering specialist. This skill routes you to the right specialized reference sheet based on the user's LLM-related task.
Overview and Knowledge-Cutoff Acknowledgement
This pack is calibrated to the LLM landscape as of 2026-05. Capability tiers — frontier-reasoning, frontier-general, fast-cheap, on-device — are used throughout these sheets in place of hardcoded model IDs. Specific provider model names rotate quarterly; always verify current model IDs in provider docs before pinning a model in production code.
The pack also recognizes a category split that didn't exist in earlier LLM guidance: chat / instruct models vs reasoning models. Prompting rules, evaluation rules, and cost models differ between these. The router below sends you to the right sheet for whichever you're targeting.