diff --git a/.gitattributes b/.gitattributes
index ad6c5330b1c..bdf96bc91bd 100644
--- a/.gitattributes
+++ b/.gitattributes
@@ -15295,5 +15295,6 @@ tools/formats.txt -text
 tools/guilds.xlsx -text
 tools/mtg-data.txt svneol=native#text/plain
 tools/mtgdata-sets-to-forge.txt svneol=native#text/plain
+tools/oracleScraper.py -text
 tools/oracleScript.py svneol=native#text/x-python
 tools/packdata.xlsx -text
diff --git a/tools/oracleScraper.py b/tools/oracleScraper.py
new file mode 100644
index 00000000000..a92b46e1b13
--- /dev/null
+++ b/tools/oracleScraper.py
@@ -0,0 +1,42 @@
+import os
+import requests
+
+setName = 'ths'
+nameStart = '<span style="font-size: 1.2em;">'
+oracleStart = '<p class="ctext"><b>'
+oracleEnd = '</b></p>'
+
+def normalizeName(name):
+	return name.lower().replace(',','').replace("'","").replace(' ', '_')
+
+def normalizeOracle(oracle):
+	return oracle.replace(u'\u2014', '-').replace(u'\u2018', "'")
+
+r = requests.get('http://magiccards.info/query?v=spoiler&s=issue&q=++e:%s/en' % setName)
+spl = r.text.split(nameStart)
+spl.pop(0) # Get rid of all of the html that comes before our first card
+
+for s in spl:
+	# Extract name and oracle from magiccards.info
+	name = s[1 + s.find(">"):s.find("</a>")]
+	oracle = s[len(oracleStart)+s.find(oracleStart):s.find(oracleEnd)].replace('<br><br>', '\\n')
+	norm = normalizeName(name)
+	# Open relative cardsfolder
+	path = os.path.join('..','res','cardsfolder', norm[0], norm+'.txt')
+
+	hasOracle = False
+	try:
+		with open(path, 'r') as f:
+			for line in f.readlines():
+				hasOracle |= line.startswith("Oracle:")
+
+		if not hasOracle:
+			with open(path, "a") as f:
+				f.write('\n')
+				f.write(normalizeOracle(oracle))
+			print '+ ',  norm
+		else:
+			print '= ', norm
+
+	except:
+		print '? ', norm
\ No newline at end of file