What problem does it solve? Choosing between AI models for a given prompt or skill is usually guesswork. This Skill replaces vibes with data by running the same prompt through Claude, GPT (via Codex CLI), and Gemini side-by-side, then reporting latency, token usage, cost, and optional quality scores. ## Core Features & Use Cases - Cross-Model Comparison: Executes one prompt across multiple providers and tabulates latency, tokens, and per-provider cost. - Optional LLM Judge: Uses Anthropic's Claude to score each output on a 0-10 quality scale for a small additional cost. - Auth Pre-Flight Check: A dry-run shows which providers are authenticated before any paid API calls are made. - Baseline Saving: Stores results as dated JSON files so future runs can be diffed to catch model quality regressions. - Use Case: You want to know which model handles your /review skill best. Run the benchmark against that skill's SKILL.md, compare the table, and save the baseline to re-check after providers update their models. ## Quick Start Ask the assistant to run a cross-model benchmark comparing Claude, GPT, and Gemini on one of your gstack skills.
