Cyber Catastrophe Bond Trigger & Portfolio Accumulation Monitor with Mastra AI
Managing accumulation risk in Cyber Catastrophe Bonds requires real-time aggregation of widespread cloud outage events across global insured portfolios. You will build an autonomous cyber catastrophe monitoring system using Mastra AI in TypeScript to track systemic cloud downtime signals against bond trigger thresholds. The system must process telemetry with sub-1 second latency and zero unparsed events.
What you are building
The core problem, expected build, and operating context for this challenge.
Build a real-time cyber catastrophe bond accumulation and trigger monitoring pipeline in TypeScript using Mastra AI workflows and agent tools.
How work is evaluated
Evaluates Mastra AI cyber cat bond monitoring workflow on cloud downtime telemetry streams.
Challenge datasets
Public datasets, plus the private datasets you have access to.
How submissions are scored
These dimensions define what the evaluator checks and which criteria separate a passable run from a strong one.
loss_positive
Estimated loss must be positive float.
This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.
workflow_execution_success
Percentage of events processed cleanly without workflow failure. • target: 1 • range: 0-1
This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.
What you should walk away with
Define Mastra AI workflows (`@mastra/core/workflows`) with structured steps for cyber incident telemetry.
Build custom Mastra tools that query cloud availability metrics and cloud outage logs.
Implement real-time aggregation models for estimating industry economic loss from hyperscaler downtime.
Output automated alerts for Cyber Cat Bond triggers when estimated claims breach attachment thresholds.
Reference links and supporting material
Synthetic stream of 150 cloud outage incidents detailing downtime durations, affected cloud services, and global user impacts.
Challenge setup
Evaluates Mastra AI cyber cat bond monitoring workflow on cloud downtime telemetry streams.
Challenge input
JSON containing event_id, cloud_provider, downtime_hours, affected_regions, portfolio_market_share
Mastra AI
Provides structured, type-safe workflow engines for serverless TypeScript applications.
Evaluated output
JSON containing total_estimated_industry_loss_usd, cat_bond_impacted (BOOLEAN), alert_level
- Estimated loss must be positive float.
- Percentage of events processed cleanly without workflow failure. • target: 1 • range: 0-1
- Benchmark: MastraCyberBench-2026
- Workflow Execution Success target: 1
- 1 public reference case
- JavaScript execution harness
- javascript sandbox (unavailable on Versalist)
Summary shows the first task, up to three tools, and up to three evidence items. Open the technical recipe to view all configured items.
View technical recipe
Configured tools
- Mastra AI · Required
- RunPod · Optional
- ERNIE 4.0 · Optional
- Mastra AI · Required
Evaluation contract
- loss_positive · Weight 1
- workflow_execution_success · Weight 1
Metrics
- workflow_execution_success · Target 1
Benchmarks
- MastraCyberBench-2026
- security
Reference cases
- 1 public reference case
Workflow tasks
- Task 1
Input: JSON containing event_id, cloud_provider, downtime_hours, affected_regions, portfolio_market_share
Output: JSON containing total_estimated_industry_loss_usd, cat_bond_impacted (BOOLEAN), alert_level
Execution harness: JavaScript
Evaluation tools
- Mastra AI · TypeScript Agent & Workflow Framework
Recipe freeze marker
No recipe freeze time is recorded.