diff --git a/weboob/browser/adapters.py b/weboob/browser/adapters.py
new file mode 100644
index 0000000000000000000000000000000000000000..60e0f50cbf305106fa000e711dd162923a874184
--- /dev/null
+++ b/weboob/browser/adapters.py
@@ -0,0 +1,41 @@
+# -*- coding: utf-8 -*-
+
+# Copyright(C) 2019 Romain Bignon
+#
+# This file is part of weboob.
+#
+# weboob is free software: you can redistribute it and/or modify
+# it under the terms of the GNU Lesser General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+#
+# weboob is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+# GNU Lesser General Public License for more details.
+#
+# You should have received a copy of the GNU Lesser General Public License
+# along with weboob. If not, see <http://www.gnu.org/licenses/>.
+
+
+import requests
+
+
+__all__ = ['HTTPAdapter']
+
+
+class HTTPAdapter(requests.adapters.HTTPAdapter):
+    def __init__(self, *args, **kwargs):
+        self._proxy_headers = kwargs.pop('proxy_headers', {})
+        super(HTTPAdapter, self).__init__(*args, **kwargs)
+
+    def add_proxy_header(self, key, value):
+        self._proxy_headers[key] = value
+
+    def update_proxy_headers(self, headers):
+        self._proxy_headers.update(headers)
+
+    def proxy_headers(self, proxy):
+        headers = super(HTTPAdapter, self).proxy_headers(proxy)
+        headers.update(self._proxy_headers)
+        return headers
diff --git a/weboob/browser/browsers.py b/weboob/browser/browsers.py
index ed79fcbb432cc4d3a194e1dc1c15893860d09ce5..01e8b19572ef97245d21f47cb644abf0f1443f35 100644
--- a/weboob/browser/browsers.py
+++ b/weboob/browser/browsers.py
@@ -52,6 +52,7 @@
 from weboob.tools.compat import basestring, unicode, urlparse, urljoin, urlencode, parse_qsl
 from weboob.tools.json import json
 
+from .adapters import HTTPAdapter
 from .cookies import WeboobCookieJar
 from .exceptions import HTTPNotFound, ClientError, ServerError
 from .sessions import FuturesSession
@@ -87,9 +88,10 @@ class Browser(object):
     Check SSL certificates.
     """
 
-    PROXIES = None
-
     MAX_RETRIES = 2
+    """
+    Maximum retries on failed requests.
+    """
 
     MAX_WORKERS = 10
     """
@@ -116,7 +118,7 @@ def asset(cls, localfile):
             return localfile
         return os.path.join(os.path.dirname(inspect.getfile(cls)), localfile)
 
-    def __init__(self, logger=None, proxy=None, responses_dirname=None, weboob=None):
+    def __init__(self, logger=None, proxy=None, responses_dirname=None, weboob=None, proxy_headers=None):
         self.logger = getLogger('browser', logger)
         self.responses_dirname = responses_dirname
         self.responses_count = 1
@@ -126,6 +128,7 @@ def __init__(self, logger=None, proxy=None, responses_dirname=None, weboob=None)
             self.VERIFY = self.asset(self.VERIFY)
 
         self.PROXIES = proxy
+        self.proxy_headers = proxy_headers or {}
         self._setup_session(self.PROFILE)
         self.url = None
         self.response = None
@@ -215,13 +218,14 @@ def _setup_session(self, profile):
 
         # defines a max_retries. It's mandatory in case a server is not
         # handling keep alive correctly, like the proxy burp
-        adapter_kwargs = dict(max_retries=self.MAX_RETRIES)
+        adapter_kwargs = dict(max_retries=self.MAX_RETRIES,
+                              proxy_headers=self.proxy_headers)
         # set connection pool size equal to MAX_WORKERS if needed
         if self.MAX_WORKERS > requests.adapters.DEFAULT_POOLSIZE:
             adapter_kwargs.update(pool_connections=self.MAX_WORKERS,
                                   pool_maxsize=self.MAX_WORKERS)
-        session.mount('https://', requests.adapters.HTTPAdapter(**adapter_kwargs))
-        session.mount('http://', requests.adapters.HTTPAdapter(**adapter_kwargs))
+        session.mount('https://', HTTPAdapter(**adapter_kwargs))
+        session.mount('http://', HTTPAdapter(**adapter_kwargs))
 
         if self.TIMEOUT:
             session.timeout = self.TIMEOUT
diff --git a/weboob/browser/sessions.py b/weboob/browser/sessions.py
index 76aca9b7d58e31044c79c4dbd385e1dfa3268969..f6630de4791b6299c1a5aa4b1afdce71e239bc89 100644
--- a/weboob/browser/sessions.py
+++ b/weboob/browser/sessions.py
@@ -26,7 +26,7 @@
     ThreadPoolExecutor = None
 
 from requests import Session
-from requests.adapters import DEFAULT_POOLSIZE, HTTPAdapter
+from requests.adapters import DEFAULT_POOLSIZE
 from requests.compat import OrderedDict, cookielib
 from requests.cookies import RequestsCookieJar, cookiejar_from_dict
 from requests.models import PreparedRequest
@@ -34,6 +34,8 @@
 from requests.structures import CaseInsensitiveDict
 from requests.utils import get_netrc_auth
 
+from .adapters import HTTPAdapter
+
 
 def merge_hooks(request_hooks, session_hooks, dict_class=OrderedDict):
     """