Use OttoBib.com and ISBNdb.com to generate a formatted bibliographic citation from an author-title search string. More explanation here: <http://wcm1.web.rice.edu/get-citations-with-isbndb-and-ottobib.html>

blookup.py

#! /usr/bin/env python
# -*- coding: utf-8 -*-

# blookup.py
# by Caleb McDaniel <http://wcm1.web.rice.edu>

# Uses ottobib.com and isbndb.com to turn an author-title search string
# into a formatted bibliographic citation.
# Example:
#   blookup.py "blight race and reunion"
# 	>>> Blight, David W. *Race and reunion : the Civil War in American
#	memory*. Cambridge, Mass: Belknap Press of Harvard University Press, 2001.

import sys
import urllib
import urllib2

# We'll be parsing XML with BS4, so you need to install BeautifulSoup and lxml.
# e.g., `sudo pip install beautifulsoup4` and `sudo pip install lxml`
# http://www.crummy.com/software/BeautifulSoup/bs4/doc/#installing-beautiful-soup
# http://www.crummy.com/software/BeautifulSoup/bs4/doc/#parser-installation

from bs4 import BeautifulSoup

# We'll also be converting HTML output to markdown.
# For this, I use Pandoc and https://github.com/kennethreitz/pyandoc
# With some minor script tweaking, you could eliminate markdown conversion
# and just get HTML output from OttoBib. See end of script for details.

import pandoc

# SETTINGS -----------------------------------------------------------------

# We will use the isbndb.com API, so you need to create an account.
# Once logged in to isbndb.com, create an access key and enter it here.
# By default, you are limited to 500 searches a day using this key.

accesskey = "XXXXXXXX"

# OttoBib can produce output in several styles. Pick your poison.
# http://www.ottobib.com/about

style = "chicago" # mla / apa

# --------------------------------------------------------------------------

# Encode the search string for isbndb.com.
query = urllib.quote_plus(sys.argv[1])

# Search isbndb.com for string.
isbndburl = "http://isbndb.com/api/books.xml?access_key=" + accesskey + "&index1=combined&value1=" + query
xml = urllib2.urlopen(isbndburl).read()

# Use BS4 to get ISBN of top search result.
soup = BeautifulSoup(xml, "xml")
tag = soup.BookData
bookisbn = tag['isbn']

# Search ottobib.com for ISBN.
ottobiburl = "http://www.ottobib.com/isbn/" + bookisbn + "/" + style
html = urllib2.urlopen(ottobiburl).read()

# Use BS4 to get the formatted citation returned by OttoBib
osoup = BeautifulSoup(html)
for br in osoup.find_all(name='br'):
	br.decompose()
citation = osoup.find("div", class_="nine columns")

# Convert OttoBib's formatted citation from HTML to Markdown using Pandoc.
doc = pandoc.Document()
doc.html = str(citation)
citation = doc.markdown

# Print the Markdown citation, removing newlines.
print citation.replace('\n', ' ')

# Or, if you don't want Markdown, comment out lines 70-75 and uncomment 78:
# print citation 

Nice idea, I did something similar, but used the v2 api and json from the standard library. It also outputs bibtex, though it looks like OttoBib can do tht, it still relies on Beautiful Soup to parse.