The SitecoreAI Pathway 1.4 release includes the following improvements:
New capabilities
Image migration for any website — When migrating from an any-website source, you can now choose whether to include media and images in the migration. When enabled, the migration engine scrapes, transfers, and imports images from your source site into SitecoreAI. The import layer also returns the GUIDs of media items that already exist in the target environment, avoiding duplicate uploads for items that were previously migrated. This capability applies only when you select the source type Any website and is not available for Sitecore XM or XP source migrations.
Automatic Download Export Structure item creation — The Download Export Structure item is now created automatically in the Sitecore Content Editor when you click the Initialize Migration button. This removes the manual export structure procedure in local PowerShell that was previously required before you could proceed with the export structure JSON.
Per-group status in template mapping — The template mapping step now displays the AI processing status for each individual template group. This gives you clearer visibility into which groups have completed analysis, are in progress, or require attention before you proceed.
Individual URL selection for any-website migration — When you select the source type Any website, you can now supply a specific list of URLs to migrate instead of relying solely on the site's sitemap. This gives you precise control over which pages are included, without requiring a full sitemap-based crawl.
Migration metrics reset correctly on re-run — When re-running a migration with updated AI instructions, metrics, such as group counts and page statuses, now reset correctly instead of accumulating from previous runs.
Migration workflow improvements
Multi-language content migration — SitecoreAI Pathway now improves migrating content across multiple languages when migrating from Sitecore XM or XP sources. Language context is preserved throughout the extraction, mapping, and import stages, ensuring that language versions arrive correctly in your target SitecoreAI environment.
Migration guidance for the Any website source updated — The instructions and workflow guidance for migrating from any website (non-XM/XP source) have been revised to reflect the current capabilities and steps.
Dynamic placeholder mapping for partial designs — SitecoreAI Pathway now correctly maps dynamic placeholders that appear within partial design renderings during XM migration, resolving cases where these were previously omitted from the migrated output.
Placeholder key pattern accuracy — The AI component mapping engine has been updated with improved placeholder key pattern recognition, resulting in more accurate mapping of components to their correct placeholder locations in XM Cloud page templates.
Source site data coverage — The AI mapping phase now reliably processes all pages from your source site. Previously, a subset of source pages could be silently omitted from the mapping output under certain conditions; this has been resolved to ensure complete and consistent coverage across grouping, template mapping, and component mapping.
Structure export improvements (SXPSiteExtraction — v1.0.5)
Language field included in extracted output — Extracted
.jsonfiles now include alanguagekey, ensuring language context is preserved through the full migration pipeline.Deeply nested field values extracted correctly — A bug that caused some deeply nested field values to be extracted incompletely has been resolved.
Improved performance — Internal refactoring of the extraction engine results in faster extraction runs, particularly for larger sites.
Improved rerun logging — When running extraction on a site that was partially extracted previously, the tool now produces clearer log output indicating which items were skipped and why.
Package renamed — The extraction package is now distributed as
SXPSiteExtraction.zip(previouslyXMComponentExtraction-<version>.zip). The version number is included inside the package. Update any automation or documentation that references the old filename.
Reliability and error-handling improvements
Accurate migration status when all component mapping pages fail — Previously, if all pages failed during the component mapping step, the workflow status incorrectly reported as completed. This has been corrected.
Media file paths updated consistently — A fix ensures that media file paths are correctly updated across all migrated content, resolving cases where some media references pointed to incorrect locations after migration.
Session expiry handling — Users are now automatically redirected to the login page when their session expires, rather than encountering an unexplained error state.
Search now applies across all pages — The search function in the migration dashboard now searches across all records, not only those visible on the current paginated page.