The knowledge of AI Engineer 2026, on tap. Ask it something.
Your agent watched it — every sentence spoken, every line that crossed the screen, every frame. Every answer comes back with the sentence, the slide, and the second it happened.
Ask one of these
None of these is answered by one talk. The model reads the corpus to build the answer and hands back the sentence, the talk and the second it was said.
Try one of these
Keyword search over every sentence spoken, every line that crossed the screen, and the frames themselves.
in this corpus
Add this corpus to your own agent
The same corpus this page is searching, on tap for whatever you are building.
mcp endpoint
…
claude code
$ claude mcp add --transport http vidtheque …