reliability-reviewer

What problem does it solve?

The Reliability Reviewer helps engineers identify reliability gaps in code changes by focusing on how errors are handled, how retries are configured, and how timeouts, health checks, and asynchronous operations could fail in production.

Core Features & Use Cases

• Missing error handling on I/O boundaries -- HTTP calls, database queries, file operations, or message queues without proper try/catch or error callbacks.
• Retry loops with no backoff or limits -- immediate, unbounded retries that create retry storms; flags for max attempts, backoff, and jitter.
• Missing timeouts and potential cascading failures -- external calls without explicit timeouts that can hang and exhaust resources.
• Error swallowing and silent failures -- catch-and-ignore patterns that propagate misleading defaults.
• Use case: during code review, identify failure propagation paths and ensure graceful degradation.

Quick Start

Ask it to review the latest code changes to identify missing error handling, timeouts, and retry configurations that affect reliability.