An actor shows as 'maintenance' but it works fine

This is a common and frustrating situation where your actor runs successfully when you test it manually, but Apify's automated quality system still flags it with a maintenance warning. The explanation is that the maintenance flag is based on automated quality checks that evaluate different criteria than your manual tests, and these checks run independently on Apify's schedule. Here are the most likely reasons your actor shows as maintenance despite appearing to work fine. Dataset schema mismatch — your actor produces output that does not exactly match the types and fields declared in your dataset_schema.json. Your manual test might work perfectly and return useful data, but if the schema says price is a number and your actor sometimes outputs price as a string like '$19.99' instead of 19.99, the automated check flags this as a violation. This is the most common cause of 'works fine but flagged' situations. Inconsistent output on default input — Apify's quality checks run your actor with its default input. If your default input is configured differently from the input you use for manual testing, the automated runs might fail or produce different output. Make sure your actor works correctly with exactly the default input shown in your input schema. Success rate over time — even if your actor works now, the maintenance flag evaluates success rate over the past 7 days. If your actor had a bad period with multiple failures last week, the flag might still be active even though you have since fixed the issue. The flag clears once enough successful runs accumulate to bring the rolling average above the threshold. Stale builds — if your actor has not been updated in 90+ days, it may be deprioritized regardless of its current functionality. To diagnose the specific cause, use the ApifyForge Schema Validator to check for any schema mismatches you might have missed. Check the quality score breakdown in the ApifyForge dashboard, which shows the individual factors contributing to your overall score. Review the automated run logs in the Apify Console, specifically the runs triggered by Apify's system (not your manual runs) to see if they show different behavior. To resolve the flag, fix the identified issue, push a new build, run your actor successfully with the default input several times, and wait for Apify's periodic re-evaluation. The flag does not clear instantly — it typically takes several days of clean runs. For more details on the maintenance system, see the related question about why actors get flagged for maintenance.