v0.1 — open source

PromptScore

Static analysis for LLM prompts.

Analyze a prompt before you send it. Get a score, find ambiguity, surface missing structure — with references to model-specific best practices.

GitHub
npx promptscore analyze prompt.txt
PS
promptscorev0.1.0
Docs ↑

12 deterministic rules

Check length, structure, examples, output format, constraints, vague language, and more. No LLM calls, no API keys.

Model profiles

YAML profiles for Claude, GPT, and a universal baseline. Rules adjust severity and suggestions per model.

Library + CLI

Use @promptscore/core in your code or run promptscore from the terminal. Zero heavy dependencies.