How to interpret Primer-BLAST output - (Feb/09/2014 )

I have designed primers using the program Oligo 7 and I want to evaluate their specificity using primer-BLAST. However, I'm not sure how to interpret the output from BLAST and determine whether I should go through with using my primers. One of the first predicted targets is a human gene with a product length of 20 nts. Both my primers are 20 nts in length. It looks like this:

>XM_005245982.1 PREDICTED: Homo sapiens complement component 1, q subcomponent, B chain (C1QB), transcript variant X1, mRNAproduct length = 20
Forward primer 1 AGTGGTTCAGAGAGGTTAAG 20
Template         24 ..A..C.............. 43

Reverse primer 1 CTTAACCTCTCTGAACCACT 20
Template         43 ..............G..T.. 24

I don't get why the 3rd nt of my primer matches with correctly with the A, but it is considered as a mismatch since the letter is shown.

In general, what are the steps you would use to interpret this data and determine whether there is a good chance the primers will amplify an off-target gene?

I would say that the chance of you amplifying a 20bp fragment and detecting it are relatively small.  However, I think your interpretation of the results is a little off.. the dots are a match, the letters are the bases that don't pair, so in your first primer,  the sequence of the target is AGA...etc rather than AGT as in your primer.

1-2 bp can make all the difference for binding, so you may well find that these primers wouldn't amplify the short sequence.  Note that a 20 bp fragment would be the same length as the primers so there is very little chance of having it amplify to any extent.