Partly a negative answer for you: it's not necessarily a lack of anything in anarchist theory so much as the presence of theory in ML orgs. Maybe that's the same thing? I'll explain my thinking. An ML group can't generally start coming out with anti-communist shit because it would be an immediate red flag. It would be an obvious honeypot from a mile away.

ML groups that kept the line after Khrushchev's secret speech might be infiltrated but it would stand out too much to try to get them to say they've now changed their minds – especially as MLs are confident that even more evidence is now available that shows they were right about the main points all along.

It's kind of expected for anarchists to reject AES, so state department messaging blends right in. That's not to say ML orgs don't come out with outrageous shite, either, like transphobia, but it's harder to publicly denounce the fundamental principles that MLs have long agreed on.

This is why I like Sakai's talk on security. The main message is that one of the best defences is learning to spot bad politics. Technical security, etc, is important, but humans will ever be the weakest link. That weakness can be identified in bad politics, in those without consistent principles. Anarchists may have consistent principles but when one of those is 'state authority is bad' their orgs can naturally be used as a mouthpiece for criticising states the CIA doesn't like.

Edit: Indeed, it seems that many non-MLs will say such things of their own accord, so some don't even need to be infiltrated.