search
Laya model
Trends
- 1laya-guard offers local safety net for coding agents●laya-guard: A local safety net for coding agents Coding agents run shell commands all the... # ai # security # opensourc
A developer has released laya-guard, an open-source Python tool that acts as a local safety net for AI coding agents. It runs a 322-million-parameter model entirely on the user's machine to judge every shell command an agent wants to execute, blocking risky operations before they happen. The approach appeals to developers worried about autonomous agents running destructive commands, while keeping code and prompts private without cloud services.
Repos
- NandhaKishorM/laya Non-autoregressive System 1 decision engine. Typed choice, score and yes/no decisions over any text in a single forward
- ollaya-dev/ollaya Run open decision models locally: pull and serve Laya, decider, NLI and GLiClass behind a TypeSafe-compatible API. Ollam
- receptron/laya Run Laya, the open-source Jev-compatible System-1 decision model, from Node.js / TypeScript via ONNX Runtime
- mizorewww/laya-mlx Native MLX runtime for Laya typed decision models — 7–14 ms short decisions on M3 Max. No text generation, PyTorch, or c
- mizorewww/laya-coreml Local Laya typed decisions on Apple Core ML and Neural Engine. Validated ports, ~5 ms short decisions on M3 Max, reprodu
- angel291592/Intent-Router Intent compiler for AI agents — converges vague requests into typed IntentSpec contracts (probe, ask, or halt before rou