The workspace configuration area of the AiManagement module is currently incomplete. Some UI components required to manage the platform settings have not yet been implemented using the MudBlazor library.
Missing items:
- MCPs association to workspaces
- "Embedding" and "Vector Store" tabs on workspace creation and editing
3 Answer(s)
-
0
- In ABP v10.7.0-rc.5 with Blazor Web App + MudBlazor, the AI Management UI you’re looking for (Workspace → MCP-Workspace association modal and the Embedding / Vector Store tabs on create/edit) is not implemented in the open-source
abpframework/abprepository. Those screens belong to the AI Management (PRO/commercial) UI, so you won’t find the missing MudBlazor components there to “enable” via configuration. - Result: there is no ABP configuration switch you can turn on to make those MudBlazor tabs/modals appear; the only options are:
- Upgrade to a newer ABP Commercial version where MudBlazor coverage for those workspace configuration features is completed (if/when shipped), or
- Implement the UI yourself in your Blazor app by calling the AI Management HTTP APIs (workspace update + MCP association + embedder/vector-store settings), or
- If you need it immediately and it exists there, use the MVC/Razor Pages UI variant (historically, some AI Management UI pieces have landed there earlier than other UIs).
If you want to implement it yourself: build a MudBlazor dialog for MCP selection and persist the selection by calling the corresponding AI Management endpoints (workspace update / MCP assignment). Likewise, add tabs for Embedding/Vector Store fields and send them in the workspace create/update DTOs used by the module’s application services.
(Links are omitted because I can’t verify a specific public ABP OSS source location for these MudBlazor components—the missing parts are not present in the indexed OSS repository content.)
This response was generated by AI. If it didn’t solve your problem, feel free to post a new answer — our support team will jump in to help you soon.
Markdown supported.Copy, paste, or drag & drop images and files (max 100 MB per file, 100 MB total per post) - In ABP v10.7.0-rc.5 with Blazor Web App + MudBlazor, the AI Management UI you’re looking for (Workspace → MCP-Workspace association modal and the Embedding / Vector Store tabs on create/edit) is not implemented in the open-source
-
0
Just to clarify: this ticket was created on the support portal precisely as advised by info@abp.io, in order to get an official confirmation and register the demand.
Could a member of the support team please confirm whether the Embedders/Vector Storage tabs are not yet implemented in the Blazor MudBlazor UI, and if so, whether this is planned for a future release?
Thank you.
Markdown supported.Copy, paste, or drag & drop images and files (max 100 MB per file, 100 MB total per post) -
1
Hi,
The MudBlazor UI of AI Management is missing these parts. The backend already supports them, only the pages were not added. We have added the MCP server connections and the Embedding / Vector Store tabs to the MudBlazor pages, and they will be in the 10.7.0 release. Your ticket has been refunded.
Until then, you can configure them with the HTTP API, for example from the Swagger UI of your HttpApi.Host.
Embedding and Vector Store
- Get the workspace with
GET /api/ai-management/workspaces/by-name/{workspaceName}. - Copy the whole response, set the fields below, and send it to
PUT /api/ai-management/workspaces/{id}. Keep the other fields as they are. The update saves every field it gets, so a missing field such asapiKeyis saved as empty.
{ "embedderProvider": "OpenAI", "embedderModelName": "text-embedding-3-small", "embedderApiKey": "<your-api-key>", "vectorStoreProvider": "Pgvector", "vectorStoreSettings": "Host=<host>;Port=5432;Database=<database>;Username=<user>;Password=<password>" }GET /api/ai-management/workspaces/embedder-providersandGET /api/ai-management/workspaces/vector-store-providersreturn the provider names you can use.If the Pgvector database doesn't exist yet, add
Persist Security Info=trueto the connection string. Otherwise creating the database fails with "No password has been provided".MCP servers
- Get the MCP server id with
GET /api/ai-management/mcp-servers/by-name/{mcpServerName}. - Connect it with
POST /api/ai-management/workspaces/{workspaceId}/mcp-servers:
{ "mcpServerId": "<mcp-server-id>", "isEnabled": true }GET /api/ai-management/workspaces/{workspaceId}/mcp-serverslists the connected servers, andDELETE /api/ai-management/workspaces/{workspaceId}/mcp-servers/{mcpServerId}removes one.Thanks
Markdown supported.Copy, paste, or drag & drop images and files (max 100 MB per file, 100 MB total per post) - Get the workspace with