7.....c5! in the Alekhine's Defense

In the book John Shaw's playing 1. e4 Caro-Kann 1...e5 and Minor Lines he suggests the move 6.g3 after 1. e4 Nf6 2. e5 Nd5 3. d4 d6 4. Nf3 dxe5 5. Nxe5 g6 rather than the main line 6.Bc4 however doesn't 6...Bg7 7.Bg2 c5! just equalize for black. Is this an oversight or am I missing something here.

yes..8 dxc5. The knight on d5 is under threat.

8....Nb4 though threatening c2 and your knight is hanging as well

Indeed, it seems that 7...c5 is fine.

White could take everything but then Black is ok after something like this :

We could wonder why 7...c5 is different from 7...0-0 followed by 8...c5, given as "the critical test" by Shaw.

Just compare :

Now, Shaw gives 10...Qc7 11.c3 Qxc5 12.Nxf7!. While 10...Bxe5 11.Qxe5 Nxc2 12.Bh6 f6 13.Qc3 Nxa1 14.Bxf8 followed by Na3 and Rxa1 and white is better.

The point is that if black plays c5 before 0-0, white cannot play the same way (with Qe2), because 7...c5 8.dxc5 Nb4 9.Qe2? Bxe5! 10.Qxe5 Nxc2 with check! Which changes everything.

So it seems u are right!

What is surprising is that nobody in my database (on 32 games) has played this 7...c5.

Fine but you are down a pawn.You can check it on the analysis....and it does not say Nxf7+. It says c3.

Edit: Nxf7 first then c3 is a tiny bit more exact...maybe lol.

 nice game 

If there is a problem, it could be 7...c5 8.dxc5 Nb4 9.Qxd8 Kxd8 10.Bf4 : the point being to play Kd2 on Nxc2, not blocking the bishop, and now the rook on h1 is almost ready to pick black knigh going to a1. Black can run into 10...Nxc2, but then 11.Kd1 Nxa1 12.Nxf7 Ke8 13.Nxh8 Bxh8 (Na6!? and Bc8 somewhere maybe) 14.Nc3, which should be better for white.

However, 10.Bf4 g5! must be ok for black : 11.Nxf7 Ke8 12.Nxh8 gxf4...

My database has no 7....c5 either I hope there aren't many of these lines achieving easy equality in Shaw's recommendations. I don't play the Alekhine's so I can't even use this novelty.