Prove whether a prompt or model variant really won before shipping with promptstats
Run statistically sound comparisons on eval results so prompt and model changes are judged by confidence bounds, not bar-chart vibes.
Prove whether a prompt or model variant really won before shipping with promptstats
Run statistically sound comparisons on eval results so prompt and model changes are judged by confidence bounds, not bar-chart vibes.
Installation
Method 1, Agent Skill Exchange
- Install from the marketplace listing: https://agentskillexchange.com/skills/prove-whether-a-prompt-or-model-variant-really-won-before-shipping-with-promptstats/
Method 2, Git clone
git clone https://github.com/agentskillexchange/skills.git && cd skills/skills/prove-whether-a-prompt-or-model-variant-really-won-before-shipping-with-promptstats
Method 3, Download ZIP
- Download the repository ZIP and extract
skills/prove-whether-a-prompt-or-model-variant-really-won-before-shipping-with-promptstats.
Method 4, Manual copy
- Copy this skill folder into your local skills directory, then reload your agent tooling.
Method 5, Fork and sync
- Fork the repository if you want to maintain local edits while syncing upstream changes.