csoai-org avatar

Mitre Atlas Mcp

csoai-org/mitre-atlas-mcp

Connects Claude to the MITRE ATLAS framework, which catalogs real-world attack patterns against AI and ML systems. Exposes tools to query adversarial techniques, list tactics across the attack lifecycle, and look up case studies of actual AI breaches. Reach for this when you're threat modeling an ML pipeline, drafting security documentation for an AI system, or need to reference specific attack vectors like model extraction, data poisoning, or evasion attacks. Built by MEOK AI Labs with EU AI Act compliance hooks baked in. Ships as a Python package with stdio transport, so it drops into Claude Desktop or any MCP client with a single config block.

tools
3
transport
STDIO
last updated
Aug 19, 2026

Tools

3 tools

Public tool metadata for what this MCP can expose to an agent.

query_atlas_technique

No parameter schema in public metadata yet.

list_atlas_tactics

No parameter schema in public metadata yet.

case_study_lookup

No parameter schema in public metadata yet.

Sponsors

Readme

MCP Scorecard: 90/100

Mitre Atlas MCP

mcp-name: io.github.CSOAI-ORG/mitre-atlas-mcp

MITRE ATLAS (Adversarial AI) MCP

MEOK AI Labs EU AI Act License PyPI

MITRE ATLAS (adversarial AI) threat landscape MCP

Buy Starter β€” Β£29/mo

Signed attestations + unlimited audits + email support. πŸ‘‰ Subscribe at meok.ai β€” instant HMAC signing key + Stripe-managed billing.

Free tier remains MIT-licensed and zero-config. Upgrade only when you need signed compliance artefacts for audit.

MITRE ATLAS (adversarial AI) threat landscape MCP. MIT


πŸš€ Quick Start

# Install via pip
pip install mitre_atlas_mcp

View the full README on GitHub

Details

Repository
github.com/csoai-org/mitre-atlas-mcp
Package
mitre-atlas-mcp
Transport
STDIO
Registry
MCP registry, active

More from csoai-org