Centralpoint Update 8.12.4
Date of Update: Monday, April 13, 2026
| Feature | Summary |
| NLS & AI > AI Providers > Embedding Providers | The AI Providers page has been enhanced with dual-track provider management. Generative (chat, summarization, record generation) and Embedding (vector-producing models for future Vector Search and RAG retrieval) are now separate, independent tracks. A site can run one provider for chat and a different one for embeddings at the same time.
Two summary chips at the top of the page display each track's current provider and model at a glance: Generative Default (blue) and Embedding Default (green).
The models list is split into two sections:
- Generative Models — LLamaSharp and OpenAI chat and summarization models, each with a "Set as Generative Default" action.
- Embedding Models — new section for vector-producing models with their dimensions shown as badges (384d, 768d, 1536d) and a "Set as Embedding Default" action. Available models include
all-MiniLM-L6-v2 (ONNX, fast local), nomic-embed-text (LLamaSharp, higher quality local), and text-embedding-3-small (OpenAI, cloud-based, requires API key).
A live Embedding Index Status panel appears below the default chips when any module is being indexed. It shows real-time progress with a green spinner, aggregate percentage, and per-module progress bars. The panel updates every few seconds without requiring a page reload.
Changing the embedding default triggers an automatic re-embedding of every enrolled module. The confirmation dialog warns that re-embedding can take hours or days on production sites and displays an estimated duration based on record count and chosen provider. Once confirmed, a processing overlay keeps the admin informed during the scrub, then the Embedding Index Status panel takes over to show live progress.
Module names shown throughout the panel now honor site aliases and navigation labels — admins see the name they set (for example, "Knowledge Base") rather than the system name (like "GenericEnhancedM").
Note: Embedding providers are configured now in preparation for future Vector Search and Knowledge Assistant features, which will query indexed content using semantic similarity rather than keyword matching. Setting the embedding default today populates the index that those future features will use. |
| Console > CAD Files > Files | CAD Files — Autodesk Integration Update
The CAD Files module has been updated to use Autodesk's current upload workflow. Autodesk retired their legacy upload endpoint, which was causing new CAD file uploads to fail. This update keeps processing working going forward — no action is required for normal day-to-day use.
New record attribute: "Regenerate CAD Base 64"
Each CAD Files record now has a "Regenerate CAD Base 64" attribute (Yes/No, defaults to No). It is only needed after your Autodesk API credentials (Client ID or Client Secret) are rotated in the CAD Files module properties.
Why this matters: when Autodesk credentials change, previously uploaded files remain tied to the old credentials at Autodesk and can no longer be displayed in the viewer. Affected records will fail to render or show an authentication error.
To recover a record after a credential change:
1. Edit the CAD Files record
2. Set "Regenerate CAD Base 64" to Yes
3. Save
The record will be re-uploaded and re-translated under the new credentials, and the attribute resets itself to No automatically. Applies only to 3D files; 2D files (.gp4, .dwg) are unaffected. |
| Console > Data Transfer > Script Builder > FileDownload | The FileDownload script in Data Transfer now supports filename replacement options. Four new Script Builder parameters have been added:- Prefix - Text to prepend to filenames (for uniqueness across modules)
- Replace Text - Text or regex pattern to find in the filename
- Replace With - Replacement text (leave empty to remove matches)
- Replace Text: Regex Enabled - Enable regular expression matching
Security hardening: ReplaceWith values containing path separators (/, \, ..) are rejected to prevent path traversal. Filenames are sanitized before and after replacement using Path.GetInvalidFileNameChars(). Regex patterns are compiled once with RegexOptions.Compiled and a 250ms timeout to prevent ReDoS attacks. Invalid regex patterns and timeouts are caught and reported via DataTransfer.Status.
The replacement logic is also applied in the Content-Disposition retry path for consistency when the primary download fails. |
| Console > Data Transfer > Taxonomy Script > Trim Path | A new Trim Path option has been added to the Data Transfer Taxonomy script when using the
Generate action. When enabled, leading and trailing delimiters are automatically removed from taxonomy paths
before processing. This is useful when importing taxonomy hierarchies from external sources like SharePoint where
paths may start or end with the delimiter character (e.g., /Zoo Trust/Surrogate Trustees/). The option appears
in the Script Builder when selecting the Taxonomy script with Action set to Generate. The existing behavior is
preserved when Trim Path is disabled (the default). |
See Full History of All Updates