What it is
A Pi extension and library that gives the agent and other extensions one consented, key-managed TypeSafe client with a batched `typesafe_evaluate` tool and offline-testable transport; requests are billable and opt-in per user.
How it uses Jev
Jev answers Choice, Score, and Noul questions about a JSON state. The `typesafe_evaluate` tool sends 1–32 questions in one batched call and returns answers, model, token usage, and elapsed time. Results are used by the agent or extension code to branch on calibrated probabilities.
Primitives:choicescorenoul
Technique worth stealing
Batch independent judgments as separate questions in one request; they run in parallel and cannot see each other.
Try it
Run `pi install npm:pi-typesafe`, then inside Pi use `/typesafe login`, `/typesafe test`, and `/typesafe enable`.
Evidence
Each line is one question put to Jev about the README. ≥ 0.60 reads as yes, ≤ 0.40 as no; in between Jev is not making a call.
- Jev-centricyes0.88
- Shows a System One patternyes0.86
- Handles uncertaintyno0.05
- Measuredno0.10
- Runnableyes0.96
- Worth recommendingunclear0.48
- Model replicano0.03
- Problem scopescore on a 0–2 scale1.05
- About Jevyes0.98
Signals by Jev jev-1.13.0, card written by DeepSeek V4.1 Flash from the README on 21 Sept 2026.