SentinelleIA · Standards & compliance

Frameworks & compliance

SentinelleIA speaks the language of agentic AI security standards and covers the EU AI Act's compliance requirements.

OWASP Agentic 2026 MAESTRO EU AI Act art. 9-15

Mapped to the seven MAESTRO layers

MAESTRO is the Cloud Security Alliance threat model for agentic AI systems, published in July 2025. It decomposes an agentic system into seven layers, from the model itself up to the ecosystem of agents that call each other. Nine agents and a separate supervisor operate across all seven.

LayerMAESTRO domainWhat the layer coversSentinelleIA coverage
L1Foundation ModelsModel level risks: jailbreaks, prompt level manipulation, unsafe or manipulated outputs.Prompt Guard, LLM Security
L2Data OperationsData pipelines, retrieval sources, third party datasets and model dependencies.Supply Chain, Tool Protection
L3Agent FrameworksThe agent runtime itself: reasoning loop, tool invocation, mandate enforcement.Supervisor (authorization boundary), AI Firewall
L4Deployment InfrastructureWhere agents run and how they reach the outside world.Gateway, Tool Protection
L5Evaluation and ObservabilityMeasurement, telemetry, detection of drift and of blind spots.Visibility, metrics and alerting chain
L6Security and ComplianceControls, evidence production, regulatory alignment.Governance, SHA-256 chained audit log
L7Agent EcosystemMulti agent interaction, delegation, cascade failure between agents.Resilience, supervisor

Two components are cross cutting and therefore appear on more than one layer: tool protection on L2 and L4, and the supervisor on L3, L6 and L7. Every one of the nine agents is placed, none is orphaned. Covering a layer means having a control that operates there at runtime, not exhausting every risk the layer contains.

Get the full mapping as a PDF →

The frameworks we cover

SentinelleIA aligns with the leading agentic AI security standards and the regulatory requirements in force.

Agentic
OWASP Top 10 for Agentic Applications2026

The reference framework for risks specific to autonomous AI agents.

OWASP
OWASP LLM Top 102025

Application-level risks of large language models.

Cloud Security Alliance
MAESTRO

Layered threat modeling for agentic systems.

MITRE
MITRE ATLAS

Adversarial tactics and techniques against ML systems.

Regulation
EU AI Actart. 9-15

Native compliance engine and enforceable proof for high-risk systems.

OWASP
OWASP AISVS

Security verification standard for AI systems.

NIST
NIST AI RMF

AI risk management, GenAI and agentic profiles.

Sovereignty
ANSSI

Security recommendations for generative and agentic AI.

ISO/IEC
ISO/IEC 42001

AI management system.

Supported protocols

SentinelleIA observes and controls agent interoperability protocols, and adds a sovereign proof layer on top.

Anthropic · 2024

MCP

Model Context Protocol. Interception and control of tool calls at runtime.

Google · 2025

A2A

Agent-to-Agent. Visibility and control of exchanges between agents.

Apex Numeris · sovereign

Sovereign inter-agent audit

SHA-256 chained audit trail, enforceable proof native to the EU AI Act (art. 12). Designed and operated with no cloud dependency.

Your agentic system, assessed against these frameworks

You submit your environment, we return an exposure report: measured attack rate and standards coverage.

Assess my exposure