[PR #316] [MERGED] Add per-page onboarding image generation and persist images in intake cache #618

New issue

Closed

opened 2026-03-13 21:00:13 +03:00 by kerem · 0 comments

kerem commented

2026-03-13 21:00:13 +03:00

Owner

📋 Pull Request Information

Original PR: https://github.com/AJaySi/ALwrity/pull/316
Author: @AJaySi
Created: 2/4/2026
Status: ✅ Merged
Merged: 2/4/2026
Merged by: @AJaySi

Base: main ← Head: codex/add-per-page-image-generation-buttons

📝 Commits (1)

379e5f9 Add onboarding page image generation

📊 Changes

7 files changed (+298 additions, -13 deletions)

View changed files

➕ backend/api/onboarding_utils/website_automation_service.py (+31 -0)
➕ backend/services/onboarding/website_intake_service.py (+33 -0)
➕ backend/services/website_automation_service.py (+42 -0)
📝 frontend/src/components/OnboardingWizard/BusinessDescriptionStep.tsx (+2 -2)
📝 frontend/src/components/OnboardingWizard/FinalStep/FinalStep.tsx (+156 -5)
📝 frontend/src/components/OnboardingWizard/WebsiteStep.tsx (+18 -0)
📝 frontend/src/services/onboardingCache.ts (+16 -6)

📄 Description

Motivation

Enable generation and preview of AI‑generated hero images for key website pages during onboarding and ensure those images travel with the intake data for site generation and previews.
Persist page images and the website intake context in the frontend cache so downstream rendering and deployment helpers can include imagery in generated HTML/markdown.

Description

Frontend: extended frontend/src/services/onboardingCache.ts with PageImages and WebsiteIntakeCache types and persisted step2 shape; WebsiteStep.tsx now saves website/analysis/business info into the onboarding cache on continue; BusinessDescriptionStep.tsx loads cached business info with typed access.
Frontend: added per‑page image generation UI to frontend/src/components/OnboardingWizard/FinalStep/FinalStep.tsx, integrating ImageGeneratorModal, building defaultPrompt from intake, previewing generated images, and saving page_images to the onboarding cache.
Backend: added backend/services/onboarding/website_intake_service.py to normalize intake and include page_images in built site_brief, backend/services/website_automation_service.py to inject <img> tags and set og:image when rendering pages, and backend/api/onboarding_utils/website_automation_service.py to prepend page images into markdown outputs for deployment.
Misc: cleaned up unused imports in FinalStep.tsx and wired cache usage across the onboarding flow so page image generation has context.

Testing

Launched the frontend dev server with npm start to sanity‑check the UI wiring and the dev server started successfully; no automated unit/integration tests were executed.
Attempted an automated Playwright screenshot of the onboarding page, but the Playwright run timed out and the screenshot step failed. No visual artifact was produced.
No other automated tests were run as part of this rollout.

Codex Task

Note

Medium Risk
Adds new onboarding data fields and injects base64 images into generated HTML/markdown, which can impact payload size and rendering/SEO meta behavior if malformed or large.

Overview
Adds per-page image generation to the onboarding Final Step (Home/About/Contact/Products) with preview, prompt auto-fill from cached website/business analysis, and persistence of page_images into the step-2 onboarding cache.

Extends the step-2 cache shape (WebsiteIntakeCache) and updates Step 2/Business Description flows to consistently save/load website URL, analysis, business info, and page_images. On the backend, introduces helpers to propagate page_images into the normalized site_brief and to inject these images into rendered HTML (including og:image) and prepend them to generated markdown outputs.

^{Written by Cursor Bugbot for commit 379e5f954c. Configure here.}

_{🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.}

## 📋 Pull Request Information **Original PR:** https://github.com/AJaySi/ALwrity/pull/316 **Author:** [@AJaySi](https://github.com/AJaySi) **Created:** 2/4/2026 **Status:** ✅ Merged **Merged:** 2/4/2026 **Merged by:** [@AJaySi](https://github.com/AJaySi) **Base:** `main` ← **Head:** `codex/add-per-page-image-generation-buttons` --- ### 📝 Commits (1) - [`379e5f9`](https://github.com/AJaySi/ALwrity/commit/379e5f954c6a6bd3f0861c052e765ac76bf8183d) Add onboarding page image generation ### 📊 Changes **7 files changed** (+298 additions, -13 deletions) <details> <summary>View changed files</summary> ➕ `backend/api/onboarding_utils/website_automation_service.py` (+31 -0) ➕ `backend/services/onboarding/website_intake_service.py` (+33 -0) ➕ `backend/services/website_automation_service.py` (+42 -0) 📝 `frontend/src/components/OnboardingWizard/BusinessDescriptionStep.tsx` (+2 -2) 📝 `frontend/src/components/OnboardingWizard/FinalStep/FinalStep.tsx` (+156 -5) 📝 `frontend/src/components/OnboardingWizard/WebsiteStep.tsx` (+18 -0) 📝 `frontend/src/services/onboardingCache.ts` (+16 -6) </details> ### 📄 Description ### Motivation - Enable generation and preview of AI‑generated hero images for key website pages during onboarding and ensure those images travel with the intake data for site generation and previews. - Persist page images and the website intake context in the frontend cache so downstream rendering and deployment helpers can include imagery in generated HTML/markdown. ### Description - Frontend: extended `frontend/src/services/onboardingCache.ts` with `PageImages` and `WebsiteIntakeCache` types and persisted `step2` shape; `WebsiteStep.tsx` now saves website/analysis/business info into the onboarding cache on continue; `BusinessDescriptionStep.tsx` loads cached business info with typed access. - Frontend: added per‑page image generation UI to `frontend/src/components/OnboardingWizard/FinalStep/FinalStep.tsx`, integrating `ImageGeneratorModal`, building `defaultPrompt` from intake, previewing generated images, and saving `page_images` to the onboarding cache. - Backend: added `backend/services/onboarding/website_intake_service.py` to normalize intake and include `page_images` in built `site_brief`, `backend/services/website_automation_service.py` to inject `<img>` tags and set `og:image` when rendering pages, and `backend/api/onboarding_utils/website_automation_service.py` to prepend page images into markdown outputs for deployment. - Misc: cleaned up unused imports in `FinalStep.tsx` and wired cache usage across the onboarding flow so page image generation has context. ### Testing - Launched the frontend dev server with `npm start` to sanity‑check the UI wiring and the dev server started successfully; no automated unit/integration tests were executed. - Attempted an automated Playwright screenshot of the onboarding page, but the Playwright run timed out and the screenshot step failed. No visual artifact was produced. - No other automated tests were run as part of this rollout. ------ [Codex Task](https://chatgpt.com/codex/tasks/task_e_6982bfc6aa348328970d71e7e2625758)  --- > [!NOTE] > **Medium Risk** > Adds new onboarding data fields and injects base64 images into generated HTML/markdown, which can impact payload size and rendering/SEO meta behavior if malformed or large. > > **Overview** > Adds per-page image generation to the onboarding *Final Step* (Home/About/Contact/Products) with preview, prompt auto-fill from cached website/business analysis, and persistence of `page_images` into the step-2 onboarding cache. > > Extends the step-2 cache shape (`WebsiteIntakeCache`) and updates Step 2/Business Description flows to consistently save/load website URL, analysis, business info, and `page_images`. On the backend, introduces helpers to propagate `page_images` into the normalized `site_brief` and to inject these images into rendered HTML (including `og:image`) and prepend them to generated markdown outputs. > > <sup>Written by [Cursor Bugbot](https://cursor.com/dashboard?tab=bugbot) for commit 379e5f954c6a6bd3f0861c052e765ac76bf8183d. Configure [here](https://cursor.com/dashboard?tab=bugbot).</sup>  --- <sub>🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.</sub>

kerem

2026-03-13 21:00:13 +03:00

closed this issue
added the
pull-request
label

No milestone

No project

No assignees

1 participant

Notifications

Due date

The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference

starred/ALwrity#618

No description provided.

Rows
Columns