Riscly docs

Guides

Reliability & simulation

Beyond security, Riscly scores the reliability of your architecture and can simulate failures to estimate business impact before they happen in production.

Reliability analysis

Send an architecture graph to get a reliability score, weighted findings (single points of failure, missing redundancy, no backups, etc.) and prioritized recommendations with an estimated risk reduction.

const report = await riscly("/analyze/reliability", {
  method: "POST",
  body: JSON.stringify({ graph }),
}).then((r) => r.json());

console.log(report.score);
for (const r of report.recommendations) {
  console.log(r.priority, r.title, `-${r.riskReductionPct}% risk`);
}

Business context

Optional context about the size of the business behind the system. The Riscly UI does not show money estimates — an outage cost derived from a graph would be a guess presented as a fact — but the /analyze endpoints will return an impact estimate if you supply this and want to compute your own.

await riscly(`/projects/${projectId}/business`, {
  method: "PUT",
  body: JSON.stringify({
    monthlyRevenue: 120000,
    activeUsers: 8000,
    currency: "EUR",
  }),
});

Auto-fill from Stripe

If Stripe is connected, fetch a live MRR / active-user suggestion from /projects/{id}/business/stripe-suggestion and pre-fill the form.

Failure simulation

Simulate an outage of a component or dependency over a time window to see the blast radius, the affected components and the estimated downtime.

const result = await riscly("/analyze/simulate", {
  method: "POST",
  body: JSON.stringify({
    graph,
    type: "node-outage",
    params: { nodeId: "db-primary" },
    durationHours: 2,
  }),
}).then((r) => r.json());

AI failure prediction

For a scanned project, get model-driven predictions of likely future failures (the model tier depends on your plan):

curl -X POST \
  https://api.riscly.com/api/projects/PROJECT_ID/scans/predict \
  -H "Authorization: Bearer $RISCLY_TOKEN" \
  -H "x-org-id: $RISCLY_ORG_ID"
PreviousRisks & one-click fixesNext Overview