https://en.wikipedia.org/wiki/Germanic_substrate_hypothesis
Based on the wiki-page about Germanic substrate hypothesis, we see that some o fthe words have their cognates in Albanian:
english 'North' - in albanian 'narth' means 'I am feeling cold', or 'it is cold'
english 'sword' - in albanian 'kordha', with the same meaning
english 'shield' - in albanian 'shilte' is a hassock
english 'bow' - in albanian 'bahe' means sling, weapon of David against Goliath
Icelandic 'drekka' - in albanian 'dreka' means lunch, the portion of middle day (afternoon)
There also similarities in other fields, which I cannot write here about, f. ex. in mythology, fairy tales etc.
Those kinds of "visual/audible" comparisons of sound-alikes are fun, but do not have any scientific meaning at all, especially when the words of two particularly divergent languages/language groups are compared (because of course in that case it's much less likely that any two words that are
real cognates, and not just similarities caused by random chance, would still look almost identical after thousands of years when the two languages are mostly very different from each other in other lexical and grammatical traits).
Other than that, Albanian does have many isoglosses with Germanic, but less than with Hellenic and Balto-Slavic. But I'm not sure the examples you've given here are the best examples of such isoglosses. Similarities are not enough. Galician "nai" is a clear cognate of German "mutter" even though the two words don't look or sound much similar at all. To establish real connections between two languages, you'd have to have a look at the comparative vocabulary of the two languages and try to identify and derive regular sound correspondences between them - then you'd be certain that they aren't just random sound-alike words, but actual and probable connections between the two languages. To give you a totally hypothetical example, this is how it really must be done:
LANGUAGE X ---- probable semantic correspondence in ----- LANGUAGE Y
drako : llowa
náikeni : dííwâdh
kadravi : wollob
navúli : dobwégh
painúlo : fiidhwégha
The words above sound mostly nothing like each other, but if you analyze them as a group of words, and not just individually, you can find regular sound correspondences that make them clear counterparts to each other and make it possible to "guess" the word to a high degree in the other language by applying those "sound rules":
1) /a/ in X corresponds to /o/ in Y
2) /k/ in X corresponds to /w/ in Y
3) /n/ in X corresponds to /d/ (initialy) or /dh/ (medial) in Y
4) /i/ in X corresponds to a dropped vowel in Y
5) unstressed /o/ in X corresponds to /a/ in Y
6) /u/ in X corresponds to /we/ in Y
7) /ai/ in X corresponds to /íí/ in Y
8) /l/ in X corresponds to /gh/ in Y
9) /dr/ in X corresponds to /ll/ in Y
Of course this is a very simplified example, but that's just to demonstrate that linguistic similarities between two languages do not necessarily - or even most often - imply that the really corresponding words will sound very similar. This kind of mass comparison looking for words that look almost identical can be very misleading, because especially when they're too similar there is a big chance that it's just random chance, coincidence, which is of course always a possibility when you have a limited number of phonemes (mostly 30-40 per language) and dozens or hundreds of thousands of words.