What it is
A standalone Hermes Agent plugin that registers a jev_evaluate tool, letting the agent classify, route, verify, score, and gate using TypeSafe AI Jev via the Vercel AI Gateway.
How it uses Jev
The agent calls jev_evaluate with a state string and named questions (choice, boolean, score). Jev returns calibrated probabilities and choices in one parallel call; the plugin wraps the result in a success/error envelope for the agent.
Primitives:choicescorenoul
Technique worth stealing
Use Jev as a fast, structured judgment layer alongside the main LLM, not as a text generator.
Try it
Copy into ~/.hermes/plugins/jev, run hermes plugins enable jev, then hermes plugins doctor jev.
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-centricyes0.71
- Shows a System One patternno0.34
- Handles uncertaintyno0.06
- Measuredno0.05
- Runnableyes0.89
- Worth recommendingno0.32
- Model replicano0.04
- Problem scopescore on a 0–2 scale1.11
- About Jevyes0.98
Signals by Jev jev-1.13.0, card written by DeepSeek V4.1 Flash from the README on 20 Sept 2026.