Reva AI
Continuous adaptive authorisation platform for humans and AI agents. Policy-driven runtime access control with MCP security, intent-aware enforcement, and AWS authorization control plane.
What it does
Authorization Management Platform (AMP) that enforces, governs, and adapts access decisions across AI agents, cloud-native applications, and data in real time. Features include a Policy Control Tower for designing and governing authorisation policies at scale, a Trust Gateway for runtime enforcement across apps/APIs/agents, Policy Intelligence for explainable decisions and behaviour analytics, and a Data Fabric for unified policy context. Offers dedicated MCP Runtime Security for securing MCP server and tool invocations with centralised policy-driven authorisation. Provides an Authorization Control Plane specifically for AWS (IAM, SCPs, AVP, AgentCore).
Security relevance
Replaces static identity-only access controls with context-aware, intent-driven runtime authorisation. For AI agents specifically, it evaluates not just who is calling but what the action means in context — addressing the gap where traditional IAM fails for autonomous systems. The MCP security module secures tool invocations with policy-as-code, preventing unauthorised tool calls and data access by AI agents.
When to use it
Use when deploying AI agents that need fine-grained, context-aware access control beyond basic RBAC. Particularly relevant for organisations using MCP servers where agent-to-tool authorisation needs centralised governance. Also valuable for modernising legacy application authorisation. Enterprise platform requiring procurement.
OWASP coverage
Risks addressed — mapped to both OWASP Top 10 standards. 1 in LLM, 5 in Agentic.
The raw record
What Yuntona stores. Single source of truth — fork it on GitHub.
name: Reva AI slug: reva-ai type: Mixed category: Identity & AppSec url: https://reva.ai reviewed: 2026-04 added: 2026-04 updated: 2026-04 risks: llm: [LLM06] asi: [ASI01, ASI02, ASI04, ASI06, ASI09] complexity: Plug & Play pricing: — audience: AppSec · CISO lifecycle: [deploy] tags: [Authorization, Commercial, MCP Security, Policy-as-Code, Runtime Security]