Definition
Full definition of error handling
Production workflows always need error handling: APIs go down, data shapes change, rate limits get hit. Tiny Command provides retries (auto-handled), error branches (alternate path on failure), and catch nodes (run something on any failure). Critical for any workflow handling money, customer communication, or irreversible operations.
In practice
Error Handling examples
Error branch
Action → on success: continue. on failure: send Slack alert + log to error sheet.
Used by
Apps that exemplify error handling
See error handling in action across real integrations.
FAQ
Common questions about error handling
What's the difference between retry and error branch?
Retry repeats the same step (good for transient failures). Error branch routes to a different step (good for permanent failures like 404s).
Should I always alert on errors?
For critical workflows, yes. For high-volume non-critical workflows, alert only on error rate spikes (>1% in 5 min) to avoid alert fatigue.
