The problem is essentially you asked BLAST to do this:

C:\Program Files\NCBI\blast-2.2.29+\bin\blastp -outfmt 5 -query MVKLTSIAAGVAAIAATASATTTLAQSDERVNLVELGVYVSDIRAHLAQYYMFQAAHPTETYPVEVAEAVFNYGDF -subject HGLQELKAELDAAVLKATGRQILTLRVRLAGAQLSWLYKEATVQEVDVIPEDGAADVRVIISNSAYGKFRKLFPG

And that fails (try it directly at the command line). Rather than raw sequence via the command line, you would normally have the sequences on disk in FASTA format, and tell BLAST the filenames.

The larger question is why are you doing lots of pair wise BLAST alignments? Might another tool be better matched? For example, EMBOSS needleall? http://emboss.sourceforge.net/apps/release/6.5/emboss/apps/needleall.html

You are right in saying "I understand that each seq should be passed as individual fasta file". The reason is that you have to pass the filename of the fasta file to BLAST, rather than the sequences inside it.

I modified your code just to pass the filenames to BLAST.

from Bio.Blast.Applications import NcbiblastpCommandline
from Bio.Blast import NCBIXML

def BlastSeq():
    SC_Fasta = "sc.fsa"
    HS_Fasta = "hsap.fsa"
    blastp = "C:\\Program Files\\NCBI\\blast-2.2.29+\\bin\\blastp"

    # I would suggest saving the blast output to a file, though you do not have to do this
    blast_output = "blast_output.xml"
    NcbiblastpCommandline(blastp, query=SC_Fasta, subject=HS_Fasta, outfmt=5, out=blast_output)()
    # Now you can read the blast output from the file and process it however you want
    blast_result_records = NCBIXML.parse(open(blast_output))

BlastSeq()

Please note that now all the pairwise alignments between the sequences in "sc.fsa" and "hsap.fsa" are stored in a single blast output file.

In your original code, it seems that you intended to call blastp once for each pair of sequences. You don't need to do this, as BLAST can handle multiple sequences in -query and -subject at the same time (thanks, Peter). However, if you insist on doing this, here is the code just to show you how hard it is.

Finally, if you would like to obtain global alignment between the sequences, I would suggest using EMBOSS needle or needleall. You can also call needle in Biopython if you like. A relevant discussion on local and global alignment can be found Is It Possible To Get Near Global Alignment From Blast?.