Reference architecture for AI agents that propose actions but cannot authorize them — immutable intent capture, an inde…
Reference architecture for AI agents that propose actions but cannot authorize them — immutable intent capture, an independent Decionis policy verdict (ALLOW/ESCALATE/BLOCK), verified human approval, and a SafeExecutor that consumes a single-use intent-bound grant. An open-source AI framework/SDK by decionis, hosted on GitHub.
decionis
api
free
Others in the same category, ranked by how often they are opened.
