diff --git a/framework/paginate.py b/framework/paginate.py
new file mode 100644
index 0000000..bbe0998
--- /dev/null
+++ b/framework/paginate.py
@@ -0,0 +1,202 @@
+# Copyright 2016 The Chromium Authors. All rights reserved.
+# Use of this source code is governed by a BSD-style
+# license that can be found in the LICENSE file or at
+# https://developers.google.com/open-source/licenses/bsd
+
+"""Classes that help display pagination widgets for result sets."""
+from __future__ import print_function
+from __future__ import division
+from __future__ import absolute_import
+
+import base64
+import logging
+import hmac
+
+import ezt
+from google.protobuf import message
+
+import settings
+from framework import exceptions
+from framework import framework_helpers
+from services import secrets_svc
+from proto import secrets_pb2
+
+
+def GeneratePageToken(request_contents, start):
+  # type: (secrets_pb2.ListRequestContents, int) -> str
+  """Encrypts a List requests's contents and generates a next page token.
+
+  Args:
+    request_contents: ListRequestContents object that holds data given by the
+      request.
+    start: int start index that should be used for the subsequent request.
+
+  Returns:
+    String next_page_token that is a serialized PageTokenContents object.
+  """
+  digester = hmac.new(secrets_svc.GetPaginationKey())
+  digester.update(request_contents.SerializeToString())
+  token_contents = secrets_pb2.PageTokenContents(
+      start=start,
+      encrypted_list_request_contents=digester.digest())
+  serialized_token = token_contents.SerializeToString()
+  # Page tokens must be URL-safe strings (see aip.dev/158)
+  # and proto string fields must be utf-8 strings while
+  # `SerializeToString()` returns binary bytes contained in a str type.
+  # So we must encode with web-safe base64 format.
+  return base64.b64encode(serialized_token)
+
+
+def ValidateAndParsePageToken(token, request_contents):
+  # type: (str, secrets_pb2.ListRequestContents) -> int
+  """Returns the start index of the page if the token is valid.
+
+  Args:
+    token: String token given in a ListFoo API request.
+    request_contents: ListRequestContents object that holds data given by the
+      request.
+
+  Returns:
+    The start index that should be used when getting the requested page.
+
+  Raises:
+    PageTokenException: if the token is invalid or incorrect for the given
+      request_contents.
+  """
+  token_contents = secrets_pb2.PageTokenContents()
+  try:
+    decoded_serialized_token = base64.b64decode(token)
+    token_contents.ParseFromString(decoded_serialized_token)
+  except (message.DecodeError, TypeError):
+    raise exceptions.PageTokenException('Invalid page token.')
+
+  start = token_contents.start
+  expected_token = GeneratePageToken(request_contents, start)
+  if hmac.compare_digest(token, expected_token):
+    return start
+  raise exceptions.PageTokenException(
+      'Request parameters must match those from the previous request.')
+
+
+# If extracting items_per_page and start values from a MonorailRequest object,
+# keep in mind that mr.num and mr.GetPositiveIntParam may return different
+# values. mr.num is the result of calling mr.GetPositiveIntParam with a default
+# value.
+class VirtualPagination(object):
+  """Class to calc Prev and Next pagination links based on result counts."""
+
+  def __init__(self, total_count, items_per_page, start, list_page_url=None,
+               count_up=True, start_param_name='start', num_param_name='num',
+               max_num=None, url_params=None, project_name=None):
+    """Given 'num' and 'start' params, determine Prev and Next links.
+
+    Args:
+      total_count: total number of artifacts that satisfy the query.
+      items_per_page: number of items to display on each page, e.g., 25.
+      start: the start index of the pagination page.
+      list_page_url: URL of the web application page that is displaying
+        the list of artifacts.  Used to build the Prev and Next URLs.
+        If None, no URLs will be built.
+      count_up: if False, count down from total_count.
+      start_param_name: query string parameter name for the start value
+        of the pagination page.
+      num_param: query string parameter name for the number of items
+        to show on a pagination page.
+      max_num: optional limit on the value of the num param.  If not given,
+        settings.max_artifact_search_results_per_page is used.
+      url_params: list of (param_name, param_value) we want to keep
+        in any new urls.
+      project_name: the name of the project we are operating in.
+    """
+    self.total_count = total_count
+    self.prev_url = ''
+    self.reload_url = ''
+    self.next_url = ''
+
+    if max_num is None:
+      max_num = settings.max_artifact_search_results_per_page
+
+    self.num = items_per_page
+    self.num = min(self.num, max_num)
+
+    if count_up:
+      self.start = start or 0
+      self.last = min(self.total_count, self.start + self.num)
+      prev_start = max(0, self.start - self.num)
+      next_start = self.start + self.num
+    else:
+      self.start = start or self.total_count
+      self.last = max(0, self.start - self.num)
+      prev_start = min(self.total_count, self.start + self.num)
+      next_start = self.start - self.num
+
+    if list_page_url:
+      if project_name:
+        list_servlet_rel_url = '/p/%s%s' % (
+            project_name, list_page_url)
+      else:
+        list_servlet_rel_url = list_page_url
+
+      self.reload_url = framework_helpers.FormatURL(
+          url_params, list_servlet_rel_url,
+          **{start_param_name: self.start, num_param_name: self.num})
+
+      if prev_start != self.start:
+        self.prev_url = framework_helpers.FormatURL(
+             url_params, list_servlet_rel_url,
+            **{start_param_name: prev_start, num_param_name: self.num})
+      if ((count_up and next_start < self.total_count) or
+          (not count_up and next_start >= 1)):
+        self.next_url = framework_helpers.FormatURL(
+           url_params, list_servlet_rel_url,
+            **{start_param_name: next_start, num_param_name: self.num})
+
+    self.visible = ezt.boolean(self.last != self.start)
+
+    # Adjust indices to one-based values for display to users.
+    if count_up:
+      self.start += 1
+    else:
+      self.last += 1
+
+  def DebugString(self):
+    """Return a string that is useful in on-page debugging."""
+    return '%s - %s of %s; prev_url:%s; next_url:%s' % (
+        self.start, self.last, self.total_count, self.prev_url, self.next_url)
+
+
+class ArtifactPagination(VirtualPagination):
+  """Class to calc Prev and Next pagination links based on a results list."""
+
+  def __init__(
+      self, results, items_per_page, start, project_name, list_page_url,
+      total_count=None, limit_reached=False, skipped=0, url_params=None):
+    """Given 'num' and 'start' params, determine Prev and Next links.
+
+    Args:
+      results: a list of artifact ids that satisfy the query.
+      items_per_page: number of items to display on each page, e.g., 25.
+      start: the start index of the pagination page.
+      project_name: the name of the project we are operating in.
+      list_page_url: URL of the web application page that is displaying
+        the list of artifacts.  Used to build the Prev and Next URLs.
+      total_count: specify total result count rather than the length of results
+      limit_reached: optional boolean that indicates that more results could
+        not be fetched because a limit was reached.
+      skipped: optional int number of items that were skipped and left off the
+        front of results.
+      url_params: list of (param_name, param_value) we want to keep
+        in any new urls.
+    """
+    if total_count is None:
+      total_count = skipped + len(results)
+    super(ArtifactPagination, self).__init__(
+        total_count, items_per_page, start, list_page_url=list_page_url,
+        project_name=project_name, url_params=url_params)
+
+    self.limit_reached = ezt.boolean(limit_reached)
+    # Determine which of those results should be visible on the current page.
+    range_start = self.start - 1 - skipped
+    range_end = range_start + self.num
+    assert 0 <= range_start <= range_end
+    self.visible_results = results[range_start:range_end]