Has anyone used Azure AI Foundry to create an agent with own data

Use the Playground and then deploy as a web app.

That's as simple as it can be

https://learn.microsoft.com/en-us/azure/ai-services/openai/use-your-data-quickstart?tabs=keyless%2Ctypescript-keyless%2Cpython-new&pivots=ai-foundry-portal

If the above response helps answer your question, remember to "Accept Answer" so that others in the community facing similar issues can easily find the solution. Your contribution is highly appreciated.

hth

Marcin

Thanks, I will try it

That's as simple as it can be

I followed the tutorial and setup a new project. I am Owner in the subscription I used and still get all these authentication issues.

I will now need to assign the required roles and understand why I get these errors.

This does not seem simple.

Soren

Errors when I try to finish data connection:

The Azure AI Search resource has disabled RBAC authentication

Azure OpenAI resource system assigned managed identity miss the following roles to the Azure AI Search resource: Search Service Contributor, Search Index Data Reader

The Azure AI Search resource has disabled system assigned managed identity

The Azure OpenAI system assigned managed identity doesn't have required role assignment to the Storage account: Storage Blob Data Contributor. Adding role assignments takes about 10 minutes to take effect.

Create an Azure AI Search resource directly from the playground

If the above response helps answer your question, remember to "Accept Answer" so that others in the community facing similar issues can easily find the solution. Your contribution is highly appreciated.

hth

Marcin

Marcin,

I very much appreciate your help, but I am still not up and running.

I did create the search from the playground, but was redirected to a new tab where I setup the search service. I uploaded a simple PDF research document and I can see the search service has data:

{
  "@odata.context": "https://mindziedevfoundrysearch.search.windows.net/indexes('test2')/$metadata#docs(*)",
  "@odata.count": 8,
  "value": [
    {
      "@search.score": 1,
      "chunk_id": "5fea8c4cd31c_aHR0cHM6Ly9taW5kemllZGV2ZWxvcG1lbnQuYmxvYi5jb3JlLndpbmRvd3MubmV0L2ZpbGV1cGxvYWQtdGVzdDIvcDUxOCUyMC0lMjBUcmFjZSUyMENsdXN0ZXJpbmclMjBpbiUyMFByb2Nlc3MlMjBNaW5pbmcucGRm0_pages_7",
      "parent_id": "aHR0cHM6Ly9taW5kemllZGV2ZWxvcG1lbnQuYmxvYi5jb3JlLndpbmRvd3MubmV0L2ZpbGV1cGxvYWQtdGVzdDIvcDUxOCUyMC0lMjBUcmFjZSUyMENsdXN0ZXJpbmclMjBpbiUyMFByb2Nlc3MlMjBNaW5pbmcucGRm0",
      "content": "(BPM 2007), volume 4714 of Lecture Notes in Computer Science, pages 328–343.\nSpringer-Verlag, Berlin, 2007.\n\n6. R. W. Hamming. Error Detecting and Error Correcting Codes. Bell System Tech\nJournal, 9:147–160, 1950.\n\n7. C. Li, M. Reichert, and A. Wombacher. Discovering Process Reference Models\nfrom Process Variants Using Clustering Techniques. Technical Report TR-CTIT-\n08-30 Centre for Telematics and Information Technology, University of Twente,\nEnschede, 2008.\n\n8. R.S. Mans, M.H. Schonenberg, M. Song, W.M.P. van der Aalst, and P.J.M. Bakker.\nProcess Mining in Health Care. In L. Azevedo and A.R. Londral, editors, Interna-\ntional Conference on Health Informatics (HEALTHINF’08), pages 118–125. IEEE\nComputer Society, January 2008.\n\n9. A.K.A. de Medeiros. Genetic Process Mining. PhD thesis, Eindhoven University\nof Technology, Eindhoven, 2006.\n\n10. A. Rozinat and W.M.P. van der Aalst. Conformance checking of processes based\non monitoring real behavior. Information Systems, 33(1):64–95, 2008.\n\n11. P.-N. Tan, M. Steinbach, and V. Kumar. Introduction to Data Mining. Addison-\nWesley Longman Publishing Co., Inc, Boston, MA, USA, 2005.\n\n12. A.J.M.M. Weijters and W.M.P. van der Aalst. Rediscovering Workflow Models\nfrom Event-Based Data using Little Thumb. Integrated Computer-Aided Engi-\nneering, 10(2):151–162, 2003.\n\n\n",
      "title": "p518 - Trace Clustering in Process Mining.pdf",
      "url": "https://mindziedevelopment.blob.core.windows.net/fileupload-test2/p518%20-%20Trace%20Clustering%20in%20Process%20Mining.pdf",
      "filepath": "p518 - Trace Clustering in Process Mining.pdf"
    },

After I fixed the permissions I still get:

"The requested information is not available in the retrieved data. Please try another query or topic."

Soren