Reliability is a feature of the engine, not a plan tier
Retries, Try/Catch, dedicated error branches, and replay — on every account, because real work fails sometimes.
Automations run against the real world — an API times out, a rate limit hits, a record is briefly locked. A tool that treats failure as an afterthought hands you a broken run and a shrug. A tool that treats it as a feature hands you a bot that heals itself and a log that tells you exactly what happened.
Retry before you catch
Most failures are transient. OpFlow's Try/Catch retries a guarded branch before the catch runs — set the count and the delay — so a blip self-heals and only a real failure falls through to your error handling. Per-step retries work the same way on any individual node.
Failures get their own branch
Every node can route on error: stop the bot, continue on the regular output, or send the failure down a dedicated error branch where you notify someone, write a record, or fall back to another path. Errors are data, not dead ends.
Reliability shouldn't be the thing you pay extra for. It should be the thing you can't turn off.
See it, replay it
Every run records its per-step input, output, status, timing and error. When something breaks, you open the run, find the step, fix the cause, and replay from any point with the same data. A run-level auto-replay policy re-drives runs that failed on transient errors without you lifting a finger.
None of this is a Team-plan upgrade. It's how the engine works.