How feasible is RF design in Sky130 using open-source tools? transition frequency should be high enough in the 1.8V transistors for 2.4GHz band circuits, right? Would it be possible to tapeout an RF design using chip ignite? I'm not sure if there would be any issues with the current I/O circuits though.

(1) Yes, it's feasible, and there have been some 2.4GHz designs. . . In fact, if you check the videos from the FOSSi Latch-Up conference last weekend, the talk just before mine is about a 2.4GHz design taped out on MPW-8 or 9. (2) Yes, there are issues with the current I/O circuits. You want to use the Caravan design, make use of the bare analog pads, and construct some very minimal ESD circuits to protect the circuit without loading the pad. 2+ GHz designs on 130-180nm will necessarily have more ESD sensitivity on the high frequency outputs.

Yes, I believe some people even had some working designs on 5GHz

I went all the way to 60 GHz, and I'm in the process of treating the data of the circuits I taped out with sky130. Passive devices work really well, active devices not so much... raw measurements (with pad parasitics) show mosfet Fmax of about 50 GHz. I'll try to de-embed the access parasitics to arrive at a more sure figure, but I'd say to stay below 30GHz if you want to use the same technology. If, however, you want to go for the same range or higher I'd recommend going for IHP's sg13g2 BiCMOS technology.

@Leonardo Gomes You mention sg13g2. Taking IHPs sg13g2 130nm BiCMOS in the lab to the extreme : A Four Channel 200 Gbps Transceiver - world record at the time (2022) https://ieeexplore.ieee.org/document/10005812

Thanks @Michael Strothjohann for sharing.