ReliOp
Run reliability checks before every deploy.
The ReliOp API plugs into your CI/CD and incident workflows. Score risk before deploy. Block bad releases. Generate fixes automatically.
- Detect high-risk changes before deploy
- Map blast radius across dependencies
- Generate targeted fixes and runbooks
From config to first API response in under 60 seconds
ReliOp is self-hosted by default. Once your instance is running, every endpoint below is available on your private network.
Works from any language, any CI runner, any orchestration layer. Your data never leaves your environment.
Into the workflows you already run
Three touchpoints. Same API. No new tool to operate.
CI/CD risk gate
Every PR that touches a service config is scanned. Builds fail when the risk score breaches your threshold — before merge, not after incident.
Blast radius + runbooks
When PagerDuty, Rootly, or incident.io fire on a service, ReliOp injects downstream impact and a targeted runbook draft into the incident record.
Pattern learning
Cluster incidents by root-cause family, cost concentration, and repeat-offender services. Feed the signal back into the next risk score.
Detect. Analyze. Fix. Learn.
Each capability is a first-class surface in the API — wired together, or used independently.
Find high-risk failure points before deploy
An 8-rule reliability engine scores service configs and produces a 0–100 operational risk score on every pull request.
Map blast radius across services
Dependency graph and cascading-failure impact for every service, computed entirely from declared config — no agents, no runtime traces.
Generate targeted remediation steps
Per-service runbooks that cite the exact violation, downstream risk, and the remediation order that reduces impact fastest.
Capture incident patterns and prevent repeats
Cluster historical incidents by root-cause family, cost concentration, and temporal signature — surface the recurrences worth fixing.
How it works in your pipeline
Five steps. No new tooling to install. Every step emits JSON your existing systems already speak.
A developer pushes a service config change.
8-rule engine maps dependencies and computes blast radius.
0–100, with per-rule violations and affected services.
Merge is blocked when risk breaches your threshold.
Targeted remediation steps posted to the PR for the developer.
Simple API, fast results
- REST / JSON. Works from any language, any CI runner, any orchestration layer.
- Four endpoints cover the core flow:
/api/audit,/api/dashboard,/api/blast-radius/<service>,/api/runbook/<service>. - Self-hosted by default. Runs on your infrastructure, authenticated by what you already use.
Enterprise-ready when you are
- Private deployment. Managed inside your network perimeter, SSO-protected, tenant-isolated.
- No data leaves your environment. Local AI option keeps every inference on your infrastructure.
- Designed for regulated systems. Audit log retention and control mapping aligned to DORA, SEC 17a-4, and NYDFS Part 500.