jev-model-tokengate
An OpenAI-compatible proxy that sits between your LLM and your users. It evaluates each sliding window of tokens while the response is still streaming and cuts the stream before a violating token can reach the screen.
How it uses Jev
Goal Target Measured --- --- --- --- Zero leakage — sensitive tokens rendered 0 0 across 5 scenarios × 3 engines ✅ Schema reliability — well-formed verdicts 100% 100% (Jev noul 0–1; Claude strict tool use) ✅ Classification on the scenario set — 5/5 on all three engines ✅ Interception latency ≤ 35ms 300ms (Jev over the public internet, 277–345ms) ❌
- Languages
- JavaScript 92% · HTML 8% · Dockerfile 1%
- License
- MIT
- Last activity
- Sep 2026
- Added
- 2026-09-23
- Stars at snapshot
- 4
- Status
- Community listing
Source: community catalog. Details and metrics reflect the published community snapshot and are not a performance endorsement. Documentation details extracted from GitHub · the README.