What problem does it solve?
Running AWS Resilience Hub v2 failure mode assessments involves many API calls, asynchronous polling, and nuanced interpretation of findings by severity, achievability, and failure category. This Skill guides an AI agent through the full assessment lifecycle so findings are correctly triaged and remediated instead of misread or ignored.
Core Features & Use Cases
- Assessment Execution: Starts failure mode assessments, checks estimated cost first, and polls status until SUCCESS or FAILED with throttling-safe intervals.
- Findings Triage: Groups findings by severity (HIGH/MEDIUM/LOW), applies a priority matrix based on policy-component achievability, and maps SEEMS failure categories (SPOF, shared fate, excessive load, latency, misconfiguration) to concrete remediation actions.
- Service Functions & Reporting: Reviews AI-generated service functions, explores topology edges for blast-radius analysis, resolves findings with validation comments, and generates FAILURE_MODE reports to S3.
- Use Case: After deploying a checkout API, ask the agent to run a failure mode assessment; it starts the assessment, waits for completion, presents HIGH-severity single-point-of-failure findings first, and recommends a FIS experiment to validate the fix.
Quick Start
Ask the agent to run a failure mode assessment on your Resilience Hub service ARN and triage the findings by severity.
