When wrapping or rethrowing an error, attach the original via Error.cause and include relevant context. Discarding the cause destroys the stack trace you need to debug.
Re-throwing a new error without the cause loses the original stack trace, turning a quick fix into a multi-hour investigation.
BeforeMerge scans your pull requests against this rule and dozens more. Get actionable feedback before code ships.
Wrapping a low-level error in a domain-specific one improves messaging, but only if you keep the original. The standard Error cause option (ES2022) chains them so the full trace survives. Add identifiers (ids, operation names) as context, never secrets.
// Bad — original cause and trace are lost
try {
await db.query(sql);
} catch {
throw new Error("Failed to load orders");
}
// Good — chain the cause, add context
try {
await db.query(sql);
} catch (err) {
throw new Error(`Failed to load orders for tenant ${tenantId}`, { cause: err });
}