Keeping academic credit on work with human rights impacts but fearing repressive regime retaliations

This is a highly speculative answer - just a thought experiment.

It may be that you will never be able to publish, safely, under your own name. The secret police in such regimes may have a long reach into other countrie,s as Russia seems to have in GB.

It might be possible, however, to create a second identity for yourself that is known to only the fewest possible, most trusted, allies. It may be too late for that if you already have worked with others on these issues and they can connect you. The difficulty, of course, with a second identity is that it is necessarily not connected to your history, including degrees, employers, etc.

But academic publishing is normally by individuals and their institutions and credentials are not as important as when seeking employment. But, then, your human rights work would be attributed to a fictitious person, which I'm pretty sure is legal as long as you aren't using such a pseudonym for fraud.

I wonder if you already have connections to reputable human rights organizations (Amnesty International, ...) that can help you with legal advice on this. Certainly a human rights lawyer familiar with the issues in your home country would be worth consulting. If you work with a reputable lawyer, that person can cache the relationship between your true-name and the pseudonym, so that if it ever becomes possible to reconnect them, you have the legal means to do so.

I don't know if the US government would be of assistance, but I don't have much faith in it at the current moment. Even our own spies identities have been "outed" here for political gain.

Yes, it can be done. Some historical examples:

J. H. Hetherington and F. D. C. Willard, Physical Review Letters, 35, 1442-1444 (1975) https://doi:10.1103/PhysRevLett.35.1442

This paper has some amusing history which you can read about on F. D. C. Willard's Wikipedia page. The catch: F. D. C Willard is actually a cat.

Polymath, D. Mathematical Sciences (2014) 1: 12. https://doi.org/10.1186/s40687-014-0012-7

Again, Polymath isn't a person, but the name of a collaboration of mathematicians.

You'll note from the second example that one journal's editors objected to the use of the pseudonym, insisting that the authors use their real names, but another journal didn't mind. You can do the same. It's possible editors object to practical jokes like the FDC Willard case (if they're aware of it), but if you are protecting your name because you fear retaliation, editors are likely to be sympathetic.