diff --git a/modules/americanexpress/__init__.py b/modules/americanexpress/__init__.py
new file mode 100644
index 00000000..7302887f
--- /dev/null
+++ b/modules/americanexpress/__init__.py
@@ -0,0 +1,23 @@
+# -*- coding: utf-8 -*-
+
+# Copyright(C) 2013 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 Affero 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 Affero General Public License for more details.
+#
+# You should have received a copy of the GNU Affero General Public License
+# along with weboob. If not, see <http://www.gnu.org/licenses/>.
+
+
+from .backend import AmericanExpressBackend
+
+__all__ = ['AmericanExpressBackend']
diff --git a/modules/americanexpress/backend.py b/modules/americanexpress/backend.py
new file mode 100644
index 00000000..b647ad03
--- /dev/null
+++ b/modules/americanexpress/backend.py
@@ -0,0 +1,71 @@
+# -*- coding: utf-8 -*-
+
+# Copyright(C) 2013 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 Affero 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 Affero General Public License for more details.
+#
+# You should have received a copy of the GNU Affero General Public License
+# along with weboob. If not, see <http://www.gnu.org/licenses/>.
+
+from __future__ import with_statement
+
+from weboob.capabilities.bank import ICapBank, AccountNotFound
+from weboob.tools.backend import BaseBackend, BackendConfig
+from weboob.tools.value import ValueBackendPassword
+
+from .browser import AmericanExpressBrowser
+
+
+__all__ = ['AmericanExpressBackend']
+
+
+class AmericanExpressBackend(BaseBackend, ICapBank):
+    NAME = 'americanexpress'
+    MAINTAINER = u'Romain Bignon'
+    EMAIL = 'romain@weboob.org'
+    VERSION = '0.f'
+    DESCRIPTION = u'American Express French bank website'
+    LICENSE = 'AGPLv3+'
+    CONFIG = BackendConfig(ValueBackendPassword('login',    label='Account ID', masked=False),
+                           ValueBackendPassword('password', label='Password of account'))
+    BROWSER = AmericanExpressBrowser
+
+    def create_default_browser(self):
+        return self.create_browser(self.config['login'].get(),
+                                   self.config['password'].get())
+
+    def iter_accounts(self):
+        with self.browser:
+            for account in self.browser.get_accounts_list():
+                yield account
+
+    def get_account(self, _id):
+        with self.browser:
+            account = self.browser.get_account(_id)
+
+        if account:
+            return account
+        else:
+            raise AccountNotFound()
+
+    def iter_history(self, account):
+        with self.browser:
+            transactions = list(self.browser.get_history(account))
+            transactions.sort(key=lambda tr: tr.rdate, reverse=True)
+            return [tr for tr in transactions if not tr._is_coming]
+
+    def iter_coming(self, account):
+        with self.browser:
+            transactions = list(self.browser.get_history(account))
+            transactions.sort(key=lambda tr: tr.rdate, reverse=True)
+            return [tr for tr in transactions if tr._is_coming]
diff --git a/modules/americanexpress/browser.py b/modules/americanexpress/browser.py
new file mode 100644
index 00000000..8943a10a
--- /dev/null
+++ b/modules/americanexpress/browser.py
@@ -0,0 +1,110 @@
+# -*- coding: utf-8 -*-
+
+# Copyright(C) 2013 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 Affero 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 Affero General Public License for more details.
+#
+# You should have received a copy of the GNU Affero General Public License
+# along with weboob. If not, see <http://www.gnu.org/licenses/>.
+
+
+from urlparse import urlsplit, parse_qsl
+
+from weboob.tools.browser import BaseBrowser, BrowserIncorrectPassword
+from weboob.tools.date import LinearDateGuesser
+
+from .pages import LoginPage, AccountsPage, TransactionsPage
+
+
+__all__ = ['AmericanExpressBrowser']
+
+
+class AmericanExpressBrowser(BaseBrowser):
+    DOMAIN = 'global.americanexpress.com'
+    PROTOCOL = 'https'
+    ENCODING = 'ISO-8859-1'
+    PAGES = {'https://global.americanexpress.com/myca/logon/.*':            LoginPage,
+             'https://global.americanexpress.com/myca/intl/acctsumm/.*':    AccountsPage,
+             'https://global.americanexpress.com/myca/intl/estatement/.*':  TransactionsPage,
+            }
+
+    def is_logged(self):
+        return self.page is not None and not self.is_on_page(LoginPage)
+
+    def home(self):
+        if self.is_logged():
+            self.location(self.buildurl('/myca/intl/acctsumm/emea/accountSummary.do'))
+        else:
+            self.login()
+
+    def login(self):
+        assert isinstance(self.username, basestring)
+        assert isinstance(self.password, basestring)
+
+        if not self.is_on_page(LoginPage):
+            self.location(self.absurl('/myca/logon/emea/action?request_type=LogonHandler&DestPage=https%3A%2F%2Fglobal.americanexpress.com%2Fmyca%2Fintl%2Facctsumm%2Femea%2FaccountSummary.do%3Frequest_type%3D%26Face%3Dfr_FR%26intlink%3Dtopnavvotrecompteneligne-HPmyca&Face=fr_FR&Info=CUExpired'), no_login=True)
+
+        self.page.login(self.username, self.password)
+
+        if not self.is_logged():
+            raise BrowserIncorrectPassword()
+
+    def go_on_accounts_list(self):
+        self.select_form(name='leftnav')
+        self.form.action = self.absurl('/myca/intl/acctsumm/emea/accountSummary.do')
+        self.submit()
+
+    def get_accounts_list(self):
+        if not self.is_on_page(AccountsPage):
+            self.go_on_accounts_list()
+        return self.page.get_list()
+
+    def get_account(self, id):
+        assert isinstance(id, basestring)
+
+        l = self.get_accounts_list()
+        for a in l:
+            if a.id == id:
+                return a
+
+        return None
+
+    def get_history(self, account):
+        if not self.is_on_page(AccountsPage):
+            self.go_on_accounts_list()
+
+        url = account._link
+
+        coming = True
+        date_guesser = LinearDateGuesser()
+
+        while url is not None:
+            self.select_form(name='leftnav')
+            self.form.action = self.absurl(url)
+            self.submit()
+
+            assert self.is_on_page(TransactionsPage)
+
+            for tr in self.page.get_history(date_guesser):
+                if tr.amount > 0:
+                    coming = False
+                tr._is_coming = coming
+                yield tr
+
+            if self.page.is_last():
+                url = None
+            else:
+                v = urlsplit(url)
+                args = dict(parse_qsl(v.query))
+                args['BPIndex'] = int(args['BPIndex']) + 1
+                url = self.buildurl(v.path, **args)
diff --git a/modules/americanexpress/favicon.png b/modules/americanexpress/favicon.png
new file mode 100644
index 00000000..9409b4bd
Binary files /dev/null and b/modules/americanexpress/favicon.png differ
diff --git a/modules/americanexpress/pages.py b/modules/americanexpress/pages.py
new file mode 100644
index 00000000..f9857d47
--- /dev/null
+++ b/modules/americanexpress/pages.py
@@ -0,0 +1,129 @@
+# -*- coding: utf-8 -*-
+
+# Copyright(C) 2013 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 Affero 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 Affero General Public License for more details.
+#
+# You should have received a copy of the GNU Affero General Public License
+# along with weboob. If not, see <http://www.gnu.org/licenses/>.
+
+
+import datetime
+from decimal import Decimal
+import re
+
+from weboob.tools.browser import BasePage, BrokenPageError
+from weboob.capabilities.bank import Account
+from weboob.tools.capabilities.bank.transactions import FrenchTransaction as Transaction
+
+
+__all__ = ['LoginPage', 'AccountsPage', 'TransactionsPage']
+
+
+class LoginPage(BasePage):
+    def login(self, username, password):
+        self.browser.select_form(name='ssoform')
+        self.browser.set_all_readonly(False)
+        self.browser['UserID'] = username.encode(self.browser.ENCODING)
+        self.browser['USERID'] = username.encode(self.browser.ENCODING)
+        self.browser['Password'] = password.encode(self.browser.ENCODING)
+        self.browser['PWD'] = password.encode(self.browser.ENCODING)
+        self.browser.submit(nologin=True)
+
+
+
+class AccountsPage(BasePage):
+    def get_list(self):
+        for box in self.document.getroot().cssselect('div.roundedBox div.contentBox'):
+            a = Account()
+            a.id = self.parser.tocleanstring(box.xpath('.//tr[@id="summaryImageHeaderRow"]//div[@class="summaryTitles"]')[0])
+            a.label = self.parser.tocleanstring(box.xpath('.//span[@class="cardTitle"]')[0])
+            a.balance = Decimal('0.0')
+            coming = self.parser.tocleanstring(self.parser.select(box, 'td#colOSBalance div.summaryValues', 1))
+            a.coming = Decimal(Transaction.clean_amount(coming))
+            a.currency = a.get_currency(coming)
+            a._link = self.parser.select(box, 'div.summaryTitles a.summaryLink', 1).attrib['href']
+
+            yield a
+
+class TransactionsPage(BasePage):
+    COL_ID = 0
+    COL_DATE = 1
+    COL_DEBIT_DATE = 2
+    COL_LABEL = 3
+    COL_VALUE = -1
+
+
+    def is_last(self):
+        current = False
+        for option in self.document.xpath('//select[@id="viewPeriod"]/option'):
+            if 'selected' in option.attrib:
+                current = True
+            elif current:
+                return False
+
+        return True
+
+    def get_debit_date(self):
+        for option in self.document.xpath('//select[@id="viewPeriod"]/option'):
+            if 'selected' in option.attrib:
+                m = re.search('(\d+) ([\w\.]+) (\d{4})$', option.text.strip())
+                if m:
+                    return datetime.date(int(m.group(3)),
+                                         self.MONTHS.index(m.group(2).rstrip('.')) + 1,
+                                         int(m.group(1)))
+
+    COL_DATE = 0
+    COL_TEXT = 1
+    COL_CREDIT = -2
+    COL_DEBIT = -1
+
+    MONTHS = ['janv', u'févr', u'mars', u'avri', u'mai', u'juin', u'juil', u'août', u'sept', u'oct', u'nov', u'déc']
+
+    def get_history(self, guesser):
+        debit_date = self.get_debit_date()
+        if debit_date is not None:
+            guesser.current_date = debit_date
+
+        for tr in reversed(self.document.xpath('//div[@id="txnsSection"]//tr[@class="tableStandardText"]')):
+            cols = tr.findall('td')
+
+            t = Transaction(tr.attrib['id'])
+
+            day, month = self.parser.tocleanstring(cols[self.COL_DATE]).split(' ', 1)
+            day = int(day)
+            month = self.MONTHS.index(month.rstrip('.')) + 1
+            date = guesser.guess_date(day, month)
+
+            try:
+                detail = self.parser.select(cols[self.COL_TEXT], 'div.hiddenROC', 1)
+            except BrokenPageError:
+                pass
+            else:
+                detail.drop_tree()
+
+            raw = (' '.join([txt.strip() for txt in cols[self.COL_TEXT].itertext()])).strip()
+            credit = self.parser.tocleanstring(cols[self.COL_CREDIT])
+            debit = self.parser.tocleanstring(cols[self.COL_DEBIT])
+
+            t.date = date
+            t.rdate = date
+            t.raw = re.sub(r'[ ]+', ' ', raw)
+            t.label = re.sub('(.*?)( \d+)?  .*', r'\1', raw).strip()
+            t.set_amount(credit, debit)
+            if t.amount > 0:
+                t.type = t.TYPE_ORDER
+            else:
+                t.type = t.TYPE_CARD
+
+            yield t
diff --git a/modules/americanexpress/test.py b/modules/americanexpress/test.py
new file mode 100644
index 00000000..1bf927e0
--- /dev/null
+++ b/modules/americanexpress/test.py
@@ -0,0 +1,32 @@
+# -*- coding: utf-8 -*-
+
+# Copyright(C) 2013 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 Affero 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 Affero General Public License for more details.
+#
+# You should have received a copy of the GNU Affero General Public License
+# along with weboob. If not, see <http://www.gnu.org/licenses/>.
+
+
+from weboob.tools.test import BackendTest
+
+
+class AmericanExpressTest(BackendTest):
+    BACKEND = 'americanexpress'
+
+    def test_americanexpress(self):
+        l = list(self.backend.iter_accounts())
+
+        a = l[0]
+        list(self.backend.iter_history(a))
+        list(self.backend.iter_coming(a))