A
licenseA
qualityC
maintenanceOpen-source permission control plane for AI agents — scan, enforce, and audit every tool call with code-level policies that prompt injection can't bypass.
14
19
Business Source 1.1
No user-submitted related servers found.
Scored across 4 tools
Each tool has a unique and clear purpose: adding guardrails, probing actions, listing guards, and reviewing decisions. No overlap exists.
All tools follow the consistent 'arai_verb_noun' pattern, using snake_case throughout. Names are descriptive and predictable.
With 4 tools, the set is appropriately scoped for a guardrail management server—neither too few nor too many.
The tools cover the core lifecycle (add, check, list, review) but lack a removal or update mechanism for guardrails, which is a minor gap.