What RuyQA does

Every pull request and every AI coding agent edit gets one of four verdicts: Verified, Needs attention, Unverified or Pending. It is worked out from your test results and from which code your tests cover, so a model never decides it. A GitHub Action and CLI post it on pull requests, an MCP server lets coding agents check their own work, and RuyQA suggests which tests cover which files so a new project doesn't start empty.

Suites with nested folder hierarchy. Four-stage status workflow: DRAFT → IN_REVIEW → APPROVED → DEPRECATED. Every update versioned with full restore. Shared steps, custom fields per project, reusable templates. Milestones track releases and deadlines.

Manual execution with keyboard shortcuts. Playwright automation. When an automated test fails with what looks like a timing or locator error, RuyQA suggests a rewrite and runs it once. A test that passes after a rewrite is flagged for review and never counted as a pass. CI agent registry with heartbeat monitoring. Webhook ingestion. Cron scheduling with timezone support.

Each agent works from a live browser or from results you have recorded, and says so when there is not enough data. Workflows can chain agents, and every run is saved with its input, output and credit use. The agents — Explorer, Drift, Strategist, Flaky Manager, Visual Analyst, Bug Reporter, Perf Analyst, Data Synthesizer, Execution Optimizer, CrossBrowser.

Full bug lifecycle: NEW → TRIAGED → IN_PROGRESS → FIXED → VERIFIED → REOPENED → CLOSED. Quality gates on every release: pass rate thresholds, critical bug limits. Baselines compare test results across points in time.

Requirements with external IDs. Link test cases with COVERS, VERIFIES, or RELATES_TO. Coverage gap analysis. Record code coverage percentages with their source (JaCoCo, Codecov, Coveralls, Istanbul or GCOV) and follow the trend. Test impact analysis shows which requirements and tests a given test relates to.

Dashboard with pass rate, flaky count, automation rate, recent activity. 7-day trend, 30-day velocity, module pass rates, team performance. CSV export for the last 30 days. Full project export as JSON.

Threaded comments on test cases with per-step indexing and @mentions. Edit locks prevent concurrent conflicts. Real-time WebSocket presence. Review workflow: DRAFT → IN_REVIEW → APPROVED. Roles: OWNER, ADMIN, MEMBER, VIEWER.

Jira — push test results, create issues from failures. Slack — notify on run results, bug filings, releases. GitHub Actions, GitLab CI, Jenkins — inbound webhooks trigger runs on push, PR, or build. Korapay and Paystack for billing.

Import: CSV/XLSX test cases, TestRail JSON, Zephyr JSON, generic JSON with suite path resolution. Export: full project JSON, 30-day analytics CSV. Exports from TestRail or Zephyr can be imported directly.

Data Subject Access Request handling: ACCESS, DELETION, CORRECTION, OPT_OUT, PORTABILITY request types. Status workflow: PENDING → VERIFYING → IN_PROGRESS → COMPLETED/REJECTED. Assignment tracking.