All projects

jevshield

lgy1027/jevshield

Sub-100ms security gate for AI agent tool calls, powered by TypeSafe's Jev (System-1) decision model. Single-request Choice/Noul/Score evaluation, dual-factor blocking matrix, calibrated-confidence routing, fail-closed parsing, zero-config local fallback. LangChain-ready.

Python
Stars
5
Forks
0

Why it is listed

A framework-neutral control layer using Jev/System One for intent routing and tool-call decisions while the application retains orchestration, retries and final responses.

Related projects

  • turnstile

    Guardrails for AI agent actions. Deterministic policy, Jev semantic checks, and replayable decisions.

    Shared topics: agent-security, ai-agents, guardrails

  • jevegis

    Guardrails for LLM apps in one API call. Prompt injection, jailbreaks, leaks, unsafe content. Built on TypeSafe Jev. MIT.

    Shared topics: guardrails, llm-security

  • jev-guard

    Vets an LLM agent's tool calls through TypeSafe AI's Jev before they run — allow, block, or hold, failing safe on uncertainty.

    Shared topics: ai-safety, guardrails