ModelFuzz
ModelFuzz secures LLM agents against prompt injection. Scan for vulnerabilities, then block unsafe tool calls at the execution layer with one decorator.
ModelFuzz esta orientada a usuarios que necesitan incorporar IA a su flujo de trabajo. Lo mas relevante: model · fuzz Problem How it works Demo GitHub ↗ v0.2.1 · pip install modelfuzz Runtime Guardrails for AI Agents Intercept and block unsafe tool calls caused by prompt injection. A single poisoned document, email, or web page can hijack an LLM through indirect prompt injection, compromising LLM agent security at the source. send_email — API keys and customer data leaked to attacker@evil.com. Categorias detectadas: Agentes. Precio: Gratis / Freemium. Licencia/terminos: MIT. Fuentes verificadas: Web oficial: https://www.modelfuzz.com/ | GitHub: https://github.com/higagan/modelfuzz.
También puede interesarte.
Show HN: Release-gate – a CLI that checks AI -agent PRs for risky changes
Rapid changes are sometimes scary. While working with AI is fun, it can also cause stress because it’s impossible to keep up with all the changes. Opening Twitter in the mor...
Soluciones tecnológicas para simplificar los procesos de ensayos preclínicos.