failure-propagation-analyzer

Analyze likely cascade paths and propagation semantics in DSDS architectures. Use when estimating blast radius, diagnosing multi-hop degradation, validating containment strategies, or explaining how a

SKILL.md

GitHub

Failure Propagation Analyzer

Overview

Model how failures spread through dependencies and identify where containment should happen. Produce an explicit trigger-to-impact chain with mitigation points.

Workflow

  1. Build dependency view
  • Identify upstream/downstream relationships and critical dependencies.
  • Mark optional fallbacks and graceful degradation paths.
  1. Define trigger failure
  • Choose initiating failure type and target component.
  • Record timing and intensity assumptions.
  1. Apply propagation semantics
  • Evaluate timeouts, retries, queue growth, resource exhaustion, and split-brain risks.
  • Track each hop with expected delay and effect.
  1. Estimate blast radius
  • Identify directly affected components, second-order impacts, and user-facing outcomes.
  • Distinguish local degradation from system-wide risk.
  1. Propose containment
  • Recommend controls: bulkheads, breakers, timeouts, backpressure, and load shedding.
  • Prioritize controls by impact reduction and implementation effort.

Output format

  1. Trigger definition
  2. Propagation chain
  3. Blast radius summary
  4. Containment recommendations
  5. Confidence and assumptions

References

  • Read references/propagation-modeling-checklist.md for analysis checklist.
  • Use these repo sources:
  • canonical-catalogue/DSDS Canonical Catalogue - Failure modes & propagation semantics.csv
  • docs/04-distributed-systems-and-failures.md (Chapters 22-24)
  • planning/IMPLEMENTATION_PLAN.md (Phase 4)