ultraqa

You can automate iterative QA cycles to detect, diagnose, and fix failing tests or build issues. The process runs repeated verification, uses diagnostic agents for root-cause analysis, applies targeted code fixes, and persists state locally until tests pass or configured cycle limits are reached.

Autonomous QA cycling is the process of running repeated verification tasks for tests, builds, linters, or typechecks. It uses architect and executor agents to diagnose failures and applies precise code changes automatically, reducing manual developer toil and speeding up resolution.

Yes, you can configure maximum cycles to continuously run verification, diagnose lint or typecheck failures, attempt automated fixes, and stop safely when checks pass or a configured cycle threshold is reached.

Yes, automated test diagnosis and fixing operates without external dependencies. It relies on internal architect and executor agents to generate root-cause analysis and apply targeted fixes while tracking failures and progress in a local state file.

The cycle control mechanism includes early-exit rules for repeated failure detection and configurable maximum cycles. It automatically stops the QA process if it detects recurring failures or reaches the defined safety threshold, preventing infinite loops.

Stateful tracking persists progress and failure data to a local state file. This ensures reproducibility across QA cycles and enables early-exit rules when maximum cycles or repeated failure detection thresholds are met.