Fast yes/no verdicts for your text and JSON.
Produces yes/no probabilities, label choices, and ordinal scores for text or JSON through the jev CLI.
When to use it
Use for classification, triage, routing, screening, ranking, filtering, or bulk labeling when a decision without an explanation is sufficient.
Give it text or JSON to judge; it returns a typed decision such as a yes/no probability, label choice, or ordinal score.
What you provide

Requires Python 3.9 or later; the bundled CLI uses only the standard library.
Authenticates the default TypeSafe provider, as an alternative to the OpenRouter key. Lookup checks the environment, then $JEV_ENV_FILE, then <config dir>/.env.
This skill
TypeSafe
Sends text to TypeSafe (irreversible)
docs.typesafe.ai
Reads TypeSafe's question-design docs