Mission
Make agent autonomy useful and safe by default through enforceable, testable, auditable policies.
About Veto
Authorization for the agent era
Veto exists to make autonomous agents safe to run in production. We build an authorization kernel that evaluates every tool call before execution, so teams can ship agents without giving up control.
What we solve
Prompt instructions are not enforcement. Veto intercepts actions at runtime and applies deterministic or LLM-backed policy decisions before side effects happen.
Built by Plaw
Veto is developed by Plaw as an open and practical control layer for AI systems. The SDK is open source, local-first, and designed to work with both self-hosted and managed cloud workflows.