There's something oddly anachronistic about making a regen set based on transistors and ICs, when this is a design whose main advantage is requiring a very small number (usually one) of active components. Here's a tube-based example:

https://w5dxs.tripod.com/receiver.htm

The other thing about superregens is that they quite easily become unintentional transmitters.

There were two articles in QST on using an HP Opto Coupler (that part is now obsolete, probably are modern alternatives) to isolate the antenna form the oscillator so it could not become a transmitter. See:

Daniel Wissell, N1BYT, “The OCR Receiver,” QST, Jun 1998

and "The OCR II Receiver" QST, Sep 2000.

Since the search engines have gone AI finding things like the June 1998 issue of QST has become impossible, for me at least. Maybe you have better search fu?

The schematic of that article is reproducer in this HP/Agilent App Note:

"Overview of High Performance Analog Optocouplers" http://promvpk.ru/Catalog/Document/5097b0fb118d6519949fe8a3

QST screen shot of shcmatic:

https://www.qsl.net/va3iul/Homebrew_RF_Circuit_Design_Ideas/...

The fcc used to require certification of commercially produced superregens for this reason.

Neither of these radios appear to be superregens though.

I don’t see a schematic, but I find the 10.4 MHz oscillator confusing. A regenerative receiver with an LO is like a motorcycle with four-wheel-drive.

In case it helps, here's the schematic that came with the instructions: https://img.incoherency.co.uk/4363