Fail-open / fail-closed
What a security control does when it cannot reach a decision because it crashed, timed out or could not read its input. Fail-open lets the action proceed, which keeps work moving but leaves a gap an attacker can provoke on purpose; fail-closed refuses it, which is safe but can stop every agent at once if the control breaks. The choice should be explicit, set per surface, and made by whoever owns the risk.
How MoorAI handles it: The administrator sets it per surface (hook, model proxy, MCP gateway, Inference hooks server) in the signed policy or the root-owned config; nothing the agent can write turns a closed surface open, and each fail-closed refusal raises an alert.
See also detection vs prevention · agent hook · semantic escalation
Related terms
-
Detection vs prevention
Testing & measurement
Detection notices something and produces a record; prevention stops it from happening. They are routinely conflated in agent-security claims, and the gap…
-
Agent hook
Agents & architecture
A command an agent host runs at fixed points in its loop — before a tool call, after a tool call, when a prompt is submitted, when a session starts or…
-
Semantic escalation
MoorAI terms
A second, optional pass: when the fast deterministic detectors return nothing and the sample is still ambiguous, the text is handed to a model for a…
Defence & response
The vocabulary of running controls for agents: knowing what AI is present, cutting what it may do down to what it uses, catching misuse, and reconstructing an incident afterwards. Where MoorAI implements a term, a separate line under the definition says how; the definition itself stays vendor-neutral. To check one agent's configuration against these ideas, use the free permission right-sizer.