Back to challenges
Challenge

Mastra AI Purdue Model Data Router with XAgent Architecture

Modern OT data architectures require strict segregation across Purdue Model levels while allowing cross-layer data synthesis. Construct a TypeScript agent workflow using Mastra AI and XAgent tooling to validate and route Level 1 SCADA tags up to Level 4 Enterprise MES systems. Success is measured by achieving zero Purdue security zone violations and 100% schema validation accuracy.

EngineeringHosted by Vera
Sign in free to start
Challenge brief

What you are building

The core problem, expected build, and operating context for this challenge.

Develop a Mastra AI agent workflow integrated with XAgent execution patterns to securely route and format OT telemetry across Purdue Model boundaries.

Delivery guide

How work is evaluated

Evaluation

Evaluates data packet validation and zone enforcement across Purdue Model level definitions.

Datasets

Shared data for this challenge

Review public datasets and any private uploads tied to your build.

Evaluation rubric

How submissions are scored

These dimensions define what the evaluator checks and which criteria separate a passable run from a strong one.

Dimensions
2 scoring checks
Binary
2 pass or fail dimensions
Ordinal
0 scaled dimensions
Dimension 1

no_direct_level0_to_level4

Rejects packets attempting direct Level 0/1 to Level 4 transmission.

Binary check

This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.

Dimension 2

routing_compliance_rate

Ratio of correctly validated and routed data packets. • target: 1 • range: 0-1

Binary check

This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.

Learning goals

What you should walk away with

  • Construct a Mastra AI workflow with specialized steps for Purdue Model Level 0-4 validation.

  • Implement XAgent tool execution patterns for automated network boundary enforcement.

  • Validate schema payloads transition between SCADA, Historian, and Enterprise MES.

  • Prevent prohibited direct cross-layer communication paths using automated policy checks.

Resources and assets

Reference links and supporting material

Dataset notes

Synthesized SCADA tag logs representing level-by-level telemetry hops across an industrial chemical plant environment.

How this agent runs

Evaluates data packet validation and zone enforcement across Purdue Model level definitions.

Challenge input

JSON payload containing source_level, destination_level, and telemetry_data

Mastra AI

Required framework for building TypeScript workflows and agent steps.

XAgent

Provides structured tool execution patterns for boundary checks.

Evaluated output

JSON result with allowed, route_path, and sanitization_status

Checks for
  • Rejects packets attempting direct Level 0/1 to Level 4 transmission.
  • Ratio of correctly validated and routed data packets. • target: 1 • range: 0-1
Proof of success
  • Benchmark: PurdueArchBench-2026
  • Routing Compliance Rate target: 100%
  • 1 public reference case
Runtime evidence
  • JavaScript execution harness
  • javascript sandbox (unavailable on Versalist)
View technical recipe

Configured tools

Action Space
  • Mastra AI · Required
  • XAgent · Optional
Orchestration
  • Mastra AI · Required

Evaluation contract

  • no_direct_level0_to_level4 · Weight 1
  • routing_compliance_rate · Weight 1

Recipe state

This is a preview. The configuration can change before the evaluation recipe is locked.

Frequently Asked Questions about Mastra AI Purdue Model Data Router with XAgent Architecture