Just in case it may be useful for someone. I've reached out FFG to clarify the following rules question: "My location has exactly one clue. I'm Roland Banks and Cover Up weaknes is in my threat area with three clues on it.

1. If I use basic investigate action with Deduction commited to it, can I remove two clues from Cover Up in case of success?
2. Can Daisy Walker with Research Notes asset in play, 4 uses (evidence) on it, remove all three clues from Cover Up and the clue from the location if she succeeds Research Notes' action test by 4 (she is in the same location with Roland and exactly one clue on it)?"

Answers:

1. No. Deduction doesn’t allow you to discover clues that aren’t at that location. If your location has 1 clue at it, you can only discover 1 clue at most when you investigate it.
2. No. Like above, Research Notes won’t allow Daisy to discover clues that aren’t already on the location. But, Daisy could discover 1 clue with Research Notes to help with Cover Up.

I was trying really hard to find a neat combo with Kymani Jones's deckbuilding options and Tool Belt, since the last expansion (maybe the previous ones too, but I haven't played the game too much prior to The Scarlet Keys) seems to have hard-to-find interactions that are both quite strong and very fun to discover and play (e.g. demolishing enemies with Gray's Anatomy, Ancient Stone and Empirical Hypothesis, you can look this up in my review of Gray's Anatomy), and I think I've found something interesting. It is not necessarily tied to Kymani at all, but I think it works especially well in some of her builds.

The idea is all about the interaction between Tool Belt and Pocket Multi Tool that allows to bypass the "Limit 1 per investigator" text on the latter. First we play one of our Pocket Multi Tools and Tool Belt, then, when we have our second Pocket Multi Tool in hand, we attach the first one to Tool Belt and play the second one, which as far as I understand becomes possible because the text on the first copy is blanked, including the "Limit 1 per investigator" line. Then we can activate the first Multi Tool in our hand and once it exhausts switch it with the Multi Tool attached to the Tool Belt to get another activation during the same round. It also works with the Spring-Loaded upgrade, but you won't be able to use both Multi Tools during the same test. This neat trick, given that we have some upgrades on the Multi Tool, lets us get two Unexpected Courages every turn (or even two free copies of Lucky!). As I've mentioned, this works very well in a Kymani Jones deck that aims to utilize Chuck Fergus, Crafty and Pilfer to grab 3 clues for one action every turn, because two instances of +2 to skill value from two copies of Pocket Multi Tool should be more than sufficient to succeed by 2 on the Pilfer test to get it back at the end of the turn. Another reason why it is very good in Kymani is how Pry Bar upgrade can single-handedly protect them from nasty treacheries, bringing their already above average (for a rogue, naturally) to a confident 5 or even 7, if necessary.

There are definitely a lot of investigators that benefit greatly from having two Pocket Multi Tools in play, and I'm sure other deckbuilding enthusiasts will mention the most interesting combinations in the comments. I'll just say that interactions like this is what makes Arkham deckbuilding discoveries so exciting.

EDIT: As it was noted in the comments, this trick does not actually work :( I'll still leave the post here, hopefully to spark the spirit of discovery to find another way to get around that pesky "Limit 1 per investigator" line.

A copy of a card is defined by title. A second copy of a card is any other card that shares the same title, regardless of cardtype, text, artwork, or any other differing characteristics between the cards.

If Daisy has the bad Necronomicon in play, she can use this to search her deck for the good Necronomicon. Pretty neat!

Lola Hayes' weakness has long been part of the reason she is never played. Not only is it bad, by nature of how Lola works she is almost always her own weakness. Until multiclass came out she st any given point in time could not use half her assets, but she's still bound to the same slot pressure of everyone else.

With multiclass cards being introduced and with her weakness being substantially nerfed in latest taboo, Lola may actually soon be a viable investigator.

While usually Arkham Horror LCG is pretty good at avoiding power creep overall, or keep it at a minimum, this card has been hit full strenght by it: you cannot really justify using this in place of Deep Knowledge, when that cards costs zero and can also allow others investigators at your location to "take" your draws to draw cards from their deck instead, making that card even more versatile than Preposterous Sketches. And it doesn't even need for your location to have a clue, mind you!

This card is completely worthless if you have Innsmouth Cospiracy in your collection: about the only investigator who would run this in place of Deep Knowledge is Joe Diamond, who can reduce the cost to 0 by putting it in his hunch deck and play it without adding curses as long as he is somewhere with a clue

Even the upgrade isn't worth it, for 2 exp you reduce the cost of the card to zero: that is it. No extra skill icon, no removal of the clue clause, no extra effect: nothing. The fact you do not add curses is not worth the sheer versatility of allowing other investigators to draw, unless you are playing solo

About the only reason why I see anyone taking Preposterous Sketches over Deep Knowledge is because your team generates so many curses, you are not likely to be able to play this consistently (as you need to have 2 "free spaces" to put two curses in the Chaos Bag to pay the cost of playing the card, if you have 9 or 10 in the game you cannot play Deep Knowledge), which is honestly fairly unlikely even if you have a 4 members party all using curse-adding cards, as the high number of players means the curses will get flushed out at a quicker pace too.

So yeah, dead card: shame