def _html_in_comment(self, comment, request, response):
'''
Find HTML code in HTML comments
'''
html_in_comment = self.HTML_RE.search(comment)
if html_in_comment and \
(comment, response.get_url()) not in self._already_reported_interesting:
# There is HTML code in the comment.
comment = comment.replace('\n', '')
comment = comment.replace('\r', '')
desc = 'A comment with the string "%s" was found in: "%s".'\
' This could be interesting.'
desc = desc % (comment, response.get_url())
i = Info('HTML comment contains HTML code', desc,
response.id, self.get_name())
i.set_dc(request.get_dc())
i.set_uri(response.get_uri())
i.add_to_highlight(html_in_comment.group(0))
kb.kb.append(self, 'html_comment_hides_html', i)
om.out.information(i.get_desc())
self._already_reported_interesting.add(
(comment, response.get_url()))
def grep(self, request, response):
'''
Plugin entry point, find feeds.
:param request: The HTTP request object.
:param response: The HTTP response object
:return: None
'''
dom = response.get_dom()
uri = response.get_uri()
# In some strange cases, we fail to normalize the document
if uri not in self._already_inspected and dom is not None:
self._already_inspected.add(uri)
# Find all feed tags
element_list = self._tag_xpath(dom)
for element in element_list:
feed_tag = element.tag
feed_type = self._feed_types[feed_tag.lower()]
version = element.attrib.get('version', 'unknown')
fmt = 'The URL "%s" is a %s version %s feed.'
desc = fmt % (uri, feed_type, version)
i = Info('Content feed resource', desc, response.id,
self.get_name())
i.set_uri(uri)
i.add_to_highlight(feed_type)
self.kb_append_uniq(self, 'feeds', i, 'URL')
def _html_in_comment(self, comment, request, response):
'''
Find HTML code in HTML comments
'''
html_in_comment = self.HTML_RE.search(comment)
if html_in_comment and \
(comment, response.get_url()) not in self._already_reported_interesting:
# There is HTML code in the comment.
comment = comment.replace('\n', '')
comment = comment.replace('\r', '')
desc = 'A comment with the string "%s" was found in: "%s".'\
' This could be interesting.'
desc = desc % (comment, response.get_url())
i = Info('HTML comment contains HTML code', desc, response.id,
self.get_name())
i.set_dc(request.get_dc())
i.set_uri(response.get_uri())
i.add_to_highlight(html_in_comment.group(0))
kb.kb.append(self, 'html_comment_hides_html', i)
om.out.information(i.get_desc())
self._already_reported_interesting.add(
(comment, response.get_url()))
def analyze_document_links(self, request, response):
'''
Find session IDs in the URI and store them in the KB.
'''
try:
doc_parser = parser_cache.dpc.get_document_parser_for(response)
except:
pass
else:
parsed_refs, _ = doc_parser.get_references()
for link_uri in parsed_refs:
if self._has_sessid(link_uri) and \
response.get_url() not in self._already_reported:
# report these informations only once
self._already_reported.add(response.get_url())
desc = 'The HTML content at "%s" contains a link (%s)'\
' which holds a session id. The ID could be leaked'\
' to third party domains through the referrer'\
' header.'
desc = desc % (response.get_url(), link_uri)
# append the info object to the KB.
i = Info('Session ID in URL', desc, response.id,
self.get_name())
i.set_uri(response.get_uri())
self.kb_append(self, 'url_session', i)
break
def analyze_document_links(self, request, response):
'''
Find session IDs in the URI and store them in the KB.
'''
try:
doc_parser = parser_cache.dpc.get_document_parser_for(response)
except:
pass
else:
parsed_refs, _ = doc_parser.get_references()
for link_uri in parsed_refs:
if self._has_sessid(link_uri) and \
response.get_url() not in self._already_reported:
# report these informations only once
self._already_reported.add(response.get_url())
desc = 'The HTML content at "%s" contains a link (%s)'\
' which holds a session id. The ID could be leaked'\
' to third party domains through the referrer'\
' header.'
desc = desc % (response.get_url(), link_uri)
# append the info object to the KB.
i = Info('Session ID in URL', desc, response.id,
self.get_name())
i.set_uri(response.get_uri())
self.kb_append(self, 'url_session', i)
break
def grep(self, request, response):
'''
Plugin entry point, search for meta tags.
:param request: The HTTP request object.
:param response: The HTTP response object
:return: None
'''
uri = response.get_uri()
if not response.is_text_or_html() or uri in self._already_inspected\
or is_404(response):
return
self._already_inspected.add(uri)
try:
dp = parser_cache.dpc.get_document_parser_for(response)
except w3afException:
return
meta_tag_list = dp.get_meta_tags()
for tag in meta_tag_list:
tag_name = self._find_name(tag)
for key, val in tag.items():
for word in self.INTERESTING_WORDS:
# Check if we have something interesting
# and WHERE that thing actually is
where = content = None
if (word in key):
where = 'name'
content = key
elif (word in val):
where = 'value'
content = val
# Now... if we found something, report it =)
if where is not None:
# The atribute is interesting!
fmt = 'The URI: "%s" sent a <meta> tag with attribute'\
' %s set to "%s" which looks interesting.'
desc = fmt % (response.get_uri(), where, content)
if self.INTERESTING_WORDS.get(tag_name, None):
usage = self.INTERESTING_WORDS[tag_name]
desc += ' The tag is used for %s.' % usage
i = Info('Interesting META tag', desc, response.id,
self.get_name())
i.set_uri(response.get_uri())
i.add_to_highlight(where, content)
self.kb_append_uniq(self, 'meta_tags', i, 'URL')
def grep(self, request, response):
'''
Plugin entry point.
:param request: The HTTP request object.
:param response: The HTTP response object
:return: None, all results are saved in the kb.
'''
uri = response.get_uri()
# 501 Code is "Not Implemented" which in some cases responds with
# this in the body:
# <body><h2>HTTP/1.1 501 Not Implemented</h2></body>
# Which creates a false positive.
if response.get_code() != 501\
and response.is_text_or_html()\
and uri not in self._already_inspected:
# Don't repeat URLs
self._already_inspected.add(uri)
body_without_tags = response.get_clear_text_body()
if body_without_tags is None:
return
for match, _, _, reqres in self._multi_re.query(body_without_tags):
if reqres == 'REQUEST':
desc = 'An HTTP request was found in the HTTP body of'\
' a response.'
i = Info('HTTP Request in HTTP body', desc, response.id,
self.get_name())
i.set_uri(uri)
i.add_to_highlight(match.group(0))
kb.kb.append(self, 'request', i)
if reqres == 'RESPONSE':
desc = 'An HTTP response was found in the HTTP body of'\
' a response.'
i = Info('HTTP Response in HTTP body', desc, response.id,
self.get_name())
i.set_uri(uri)
i.add_to_highlight(match.group(0))
kb.kb.append(self, 'response', i)
def crawl(self, fuzzable_request):
"""
Find CAPTCHA images.
:param fuzzable_request: A fuzzable_request instance that contains
(among other things) the URL to test.
"""
result, captchas = self._identify_captchas(fuzzable_request)
if result:
for captcha in captchas:
desc = 'Found a CAPTCHA image at: "%s".' % captcha.img_src
response_ids = [response.id for response in captcha.http_responses]
i = Info("Captcha image detected", desc, response_ids, self.get_name())
i.set_uri(captcha.img_src)
kb.kb.append(self, "CAPTCHA", i)
om.out.information(i.get_desc())
def analyze_uri(self, request, response):
'''
Find session IDs in the URI and store them in the KB.
'''
request_uri = request.get_uri()
if self._has_sessid(request_uri) and \
response.get_url() not in self._already_reported:
# report these informations only once
self._already_reported.add(response.get_url())
desc = 'The URL "%s" contains a session id which could be'\
' leaked to third party domains through the referrer'\
' header.'
desc = desc % request_uri
# append the info object to the KB.
i = Info('Session ID in URL', desc, response.id, self.get_name())
i.set_uri(response.get_uri())
self.kb_append(self, 'url_session', i)
def analyze_uri(self, request, response):
'''
Find session IDs in the URI and store them in the KB.
'''
request_uri = request.get_uri()
if self._has_sessid(request_uri) and \
response.get_url() not in self._already_reported:
# report these informations only once
self._already_reported.add(response.get_url())
desc = 'The URL "%s" contains a session id which could be'\
' leaked to third party domains through the referrer'\
' header.'
desc = desc % request_uri
# append the info object to the KB.
i = Info('Session ID in URL', desc, response.id,
self.get_name())
i.set_uri(response.get_uri())
self.kb_append(self, 'url_session', i)
def _interesting_word(self, comment, request, response):
'''
Find interesting words in HTML comments
'''
comment = comment.lower()
for word in self._multi_in.query(response.body):
if (word, response.get_url()) not in self._already_reported_interesting:
desc = 'A comment with the string "%s" was found in: "%s".'\
' This could be interesting.'
desc = desc % (word, response.get_url())
i = Info('Interesting HTML comment', desc,
response.id, self.get_name())
i.set_dc(request.get_dc())
i.set_uri(response.get_uri())
i.add_to_highlight(word)
kb.kb.append(self, 'interesting_comments', i)
om.out.information(i.get_desc())
self._already_reported_interesting.add((word,
response.get_url()))
def crawl(self, fuzzable_request):
'''
Find CAPTCHA images.
:param fuzzable_request: A fuzzable_request instance that contains
(among other things) the URL to test.
'''
result, captchas = self._identify_captchas(fuzzable_request)
if result:
for captcha in captchas:
desc = 'Found a CAPTCHA image at: "%s".' % captcha.img_src
response_ids = [
response.id for response in captcha.http_responses
]
i = Info('Captcha image detected', desc, response_ids,
self.get_name())
i.set_uri(captcha.img_src)
kb.kb.append(self, 'CAPTCHA', i)
om.out.information(i.get_desc())
def _interesting_word(self, comment, request, response):
'''
Find interesting words in HTML comments
'''
comment = comment.lower()
for word in self._multi_in.query(response.body):
if (word, response.get_url()
) not in self._already_reported_interesting:
desc = 'A comment with the string "%s" was found in: "%s".'\
' This could be interesting.'
desc = desc % (word, response.get_url())
i = Info('Interesting HTML comment', desc, response.id,
self.get_name())
i.set_dc(request.get_dc())
i.set_uri(response.get_uri())
i.add_to_highlight(word)
kb.kb.append(self, 'interesting_comments', i)
om.out.information(i.get_desc())
self._already_reported_interesting.add(
(word, response.get_url()))
def grep(self, request, response):
'''
Plugin entry point.
:param request: The HTTP request object.
:param response: The HTTP response object
:return: None, all results are saved in the kb.
'''
try:
dp = parser_cache.dpc.get_document_parser_for(response)
except w3afException:
return
# Note:
# - With parsed_references I'm 100% that it's really something in the
# HTML that the developer intended to add.
#
# - The re_references are the result of regular expressions, which in
# some cases are just false positives.
#
parsed_references, _ = dp.get_references()
for ref in parsed_references:
qs = ref.querystring
for param_name in qs:
# This for loop is to address the repeated parameter name issue
for element_index in xrange(len(qs[param_name])):
if self._is_strange(request, param_name, qs[param_name][element_index])\
and (ref.uri2url(), param_name) not in self._already_reported:
# Don't repeat findings
self._already_reported.add((ref.uri2url(), param_name))
desc = 'The URI: "%s" has a parameter named: "%s"'\
' with value: "%s", which is very uncommon.'\
' and requires manual verification.'
desc = desc % (response.get_uri(), param_name,
qs[param_name][element_index])
i = Info('Uncommon query string parameter', desc,
response.id, self.get_name())
i.set_uri(ref)
i.set_var(param_name)
i['parameter_value'] = qs[param_name][element_index]
i.add_to_highlight(qs[param_name][element_index])
self.kb_append(self, 'strange_parameters', i)
# To find this kind of vulns
# http://thedailywtf.com/Articles/Oklahoma-
# Leaks-Tens-of-Thousands-of-Social-Security-Numbers,-Other-
# Sensitive-Data.aspx
if self._is_SQL(request, param_name, qs[param_name][element_index])\
and ref not in self._already_reported:
# Don't repeat findings
self._already_reported.add(ref)
desc = 'The URI: "%s" has a parameter named: "%s"'\
' with value: "%s", which is a SQL query.'
desc = desc % (response.get_uri(), param_name,
qs[param_name][element_index])
v = Vuln('Parameter has SQL sentence', desc,
severity.LOW, response.id, self.get_name())
v.set_uri(ref)
v.set_var(param_name)
v['parameter_value'] = qs[param_name][element_index]
v.add_to_highlight(qs[param_name][element_index])
self.kb_append(self, 'strange_parameters', v)
def grep(self, request, response):
'''
Plugin entry point.
:param request: The HTTP request object.
:param response: The HTTP response object
:return: None, all results are saved in the kb.
'''
try:
dp = parser_cache.dpc.get_document_parser_for(response)
except w3afException:
return
# Note:
# - With parsed_references I'm 100% that it's really something in the
# HTML that the developer intended to add.
#
# - The re_references are the result of regular expressions, which in
# some cases are just false positives.
#
parsed_references, _ = dp.get_references()
for ref in parsed_references:
qs = ref.querystring
for param_name in qs:
# This for loop is to address the repeated parameter name issue
for element_index in xrange(len(qs[param_name])):
if self._is_strange(request, param_name, qs[param_name][element_index])\
and (ref.uri2url(), param_name) not in self._already_reported:
# Don't repeat findings
self._already_reported.add((ref.uri2url(), param_name))
desc = 'The URI: "%s" has a parameter named: "%s"'\
' with value: "%s", which is very uncommon.'\
' and requires manual verification.'
desc = desc % (response.get_uri(), param_name,
qs[param_name][element_index])
i = Info('Uncommon query string parameter', desc,
response.id, self.get_name())
i.set_uri(ref)
i.set_var(param_name)
i['parameter_value'] = qs[param_name][element_index]
i.add_to_highlight(qs[param_name][element_index])
self.kb_append(self, 'strange_parameters', i)
# To find this kind of vulns
# http://thedailywtf.com/Articles/Oklahoma-
# Leaks-Tens-of-Thousands-of-Social-Security-Numbers,-Other-
# Sensitive-Data.aspx
if self._is_SQL(request, param_name, qs[param_name][element_index])\
and ref not in self._already_reported:
# Don't repeat findings
self._already_reported.add(ref)
desc = 'The URI: "%s" has a parameter named: "%s"'\
' with value: "%s", which is a SQL query.'
desc = desc % (response.get_uri(), param_name,
qs[param_name][element_index])
v = Vuln('Parameter has SQL sentence', desc,
severity.LOW, response.id, self.get_name())
v.set_uri(ref)
v.set_var(param_name)
v['parameter_value'] = qs[param_name][element_index]
v.add_to_highlight(qs[param_name][element_index])
self.kb_append(self, 'strange_parameters', v)
