Basedash AI Sources - Trust every AI answer by seeing what built it

Basedash AI Sources shows exactly what produced every answer. Open Sources to inspect the tables, definitions, charts, data sources, read-only MCP connections, and web pages the AI used—plus every query, its SQL, and a preview of returned rows. Completed work collapses into a compact “Analyzed for…” summary, while actions that change something appear as clear cards beneath the answer. Ask the question, then see the receipts.

Add a comment

Replies

Best
Hey everyone, Max here from Basedash. Today we're launching AI Sources: a direct way to verify every AI answer by seeing exactly what built it. Open Sources beneath an answer to inspect the tables, definitions, charts, data sources, read-only MCP connections, and web pages the assistant used. Each query includes the exact SQL and a preview of the rows it returned. Once the answer is complete, the long trail of thinking collapses into a compact “Analyzed for…” summary. Changes stay distinct from evidence. If the assistant runs a mutating query, creates a definition or automation, or updates AI context, that work appears as a clear card beneath the answer. We use Sources to check important company answers before sharing them. Instead of reconstructing the analysis, we can inspect the same rows and SQL immediately. The Product Hunt community gets an extra week on their trial this week. Happy to answer anything.

 The SQL and row previews make it much easier to see where an answer comes from. Really useful feature.

We’re very excited to launch this one! Let us know if you have any feedback or try it out.

 this idea has been on our todo list from soon after we launched chat over a year ago. Happy with how it turned out.