Align GSC SearchAnalytics requests to docs, fix decay logic, add task-report endpoints/UI and tests

[AJaySi]

Motivation

• Ensure Search Analytics payloads conform to Google Search Console documentation to avoid malformed API calls and unstable behavior.
• Fix an incorrect decay comparison that used a sliced 60-day window and caused unreliable period-over-period insights.
• Provide a single, reviewable contract for issue-task reporting used by both Onboarding Step 5 and the SEO Dashboard.
• Improve reviewer confidence by adding focused tests and a concise design-review document describing changes and rationale.

Description

• Added a docs-aligned request builder in backend/services/gsc_service.py (_build_search_analytics_request) and constants DEFAULT_SEARCH_TYPE='web' and MAX_ROW_LIMIT=25000, and switched verification/aggregate/query calls to use it to enforce type, bounded rowLimit, non-negative startRow, and omitting dimensions for aggregate totals.
• Fixed period-over-period logic in backend/services/gsc_task_report_service.py by adding _fetch_query_rows_range(...) and comparing equivalent windows (current 30d vs previous 30d) instead of slicing a 60d set, and implemented the GSCTaskReportService used to construct the 4 issue sections and Google query templates.
• Integrated the task-report surface across layers: new router endpoints in backend/routers/gsc_auth.py (GET /gsc/task-reports, POST /gsc/task-reports/run), onboarding bootstrap to create/verify platform-insights tasks in backend/api/onboarding_utils/onboarding_completion_service.py, a reusable frontend panel frontend/src/components/shared/GSCTaskReportsPanel.tsx, client methods in frontend/src/api/gsc.ts, and added the panel into IntegrationsStep.tsx and SEODashboard.tsx.
• Added a concise reviewer document docs/gsc_step5_dashboard_design_review.md summarizing design considerations, issues solved (1–4), exact code edits, and validation performed.

Testing

• Ran unit tests: PYTHONPATH=backend pytest -q backend/services/gsc_query_request_shapes_tests.py which passed (2 passed) and produced expected warnings but no failures.
• Performed Python compilation checks with python -m compileall on modified backend modules which completed successfully.

---

Codex Task

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated (UTC)
alwrity-ai	Canceled		Feb 13, 2026 6:44am

[chatgpt-codex-connector]

You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard.