Data agent inside Fabric ask question on your Business Central Data
Data Agents in Microsoft Fabric enable users to ask natural language questions about Business Central data without using APIs. The feature works with Mirrored Data sources for near real-time data synchronization and can be integrated into Copilot Studio for enhanced user experience.
post Discover Microsoft Business Central - The wonderful world of Business Central
"you can really easy create AI applications with your data and ask questions on your data"
Key points
- Fabric Data Agents answer natural language questions on Business Central data already stored in Fabric, without API calls
- Mirrored data sources synced with bc2adls give near real-time data
- Keep tables at 100 columns or fewer, or answer quality drops
- Agent instructions let it join tables, for example Customer Ledger Entries with Customer to group by City
- Querying Fabric instead of the MCP server or APIs keeps load off Business Central users, and the agent can be added to Copilot Studio
Quotes
- “you can really easy create AI applications with your data and ask questions on your data” (Explains the primary capability: using natural language to query Business Central data without technical complexity)
- “if there is an huge demand of request your Business Central users won’t affect it” (Highlights the performance benefit of querying Fabric data instead of directly calling Business Central APIs)
- “you almost get near real time data if you choose that option inside bc2adls to synchronize your date” (Shows how Mirrored Data sources provide data freshness comparable to real-time requirements)
AL objects mentioned
As named in the post. A name that matches one object page by exact type and name links to it; the others stay as named.
- table “Customer Ledger Entries”
- table 18 “Customer”
Not found in BC28-30: table “Customer Ledger Entries”.
Context
- Features: Data Agents in Fabric, Mirrored Data sources, Natural language queries on Business Central data, SQL query generation, Agent configuration instructions, Table joins via agent instructions, Copilot Studio integration
Source: Discover MS BC - Bert Verbeek, community blog. Summary, key points and quotes are derived (CONTENT-NOTICE.md); read the original for the full text.