diff --git a/mypy.ini b/mypy.ini
--- a/mypy.ini
+++ b/mypy.ini
@@ -33,9 +33,6 @@
 [mypy-pygments.*]
 ignore_missing_imports = True
 
-[mypy-pypandoc.*]
-ignore_missing_imports = True
-
 [mypy-pytest.*]
 ignore_missing_imports = True
 
diff --git a/requirements.txt b/requirements.txt
--- a/requirements.txt
+++ b/requirements.txt
@@ -15,7 +15,6 @@
 lxml
 prometheus_client
 pygments
-pypandoc
 python-dateutil
 pyyaml
 requests
diff --git a/swh/web/assets/src/bundles/webapp/webapp.css b/swh/web/assets/src/bundles/webapp/webapp.css
--- a/swh/web/assets/src/bundles/webapp/webapp.css
+++ b/swh/web/assets/src/bundles/webapp/webapp.css
@@ -442,6 +442,21 @@
     background-image: linear-gradient(to left, #f0f0f0, #8c8b8b, #f0f0f0);
 }
 
+/* Ensure that section title with link is colored like standard section title */
+.swh-readme h1 a,
+.swh-readme h2 a,
+.swh-readme h3 a,
+.swh-readme h4 a,
+.swh-readme h5 a,
+.swh-readme h6 a {
+    color: #e20026;
+}
+
+/* Make list compact in reStructuredText rendering */
+.swh-rst li p {
+    margin-bottom: 0;
+}
+
 .swh-readme-txt pre {
     background: none;
     border: none;
diff --git a/swh/web/browse/utils.py b/swh/web/browse/utils.py
--- a/swh/web/browse/utils.py
+++ b/swh/web/browse/utils.py
@@ -5,7 +5,6 @@
 
 import base64
 import magic
-import pypandoc
 import stat
 import textwrap
 
@@ -23,7 +22,7 @@
 from swh.web.common.origin_visits import get_origin_visit
 from swh.web.common.utils import (
     reverse, format_utc_iso_date, get_swh_persistent_id,
-    swh_object_icons
+    swh_object_icons, rst_to_html
 )
 from swh.web.config import get_config
 
@@ -1049,8 +1048,7 @@
         else:
             try:
                 rst_doc = request_content(readme_sha1)
-                readme_html = pypandoc.convert_text(rst_doc['raw_data'],
-                                                    'html', format='rst')
+                readme_html = rst_to_html(rst_doc['raw_data'])
                 cache.set(cache_entry_id, readme_html)
             except Exception as exc:
                 sentry_sdk.capture_exception(exc)
diff --git a/swh/web/common/swh_templatetags.py b/swh/web/common/swh_templatetags.py
--- a/swh/web/common/swh_templatetags.py
+++ b/swh/web/common/swh_templatetags.py
@@ -3,53 +3,30 @@
 # License: GNU Affero General Public License version 3, or any later version
 # See top-level LICENSE file for more information
 
-from inspect import cleandoc
 import json
 import re
 
+from inspect import cleandoc
+
 from django import template
 from django.core.serializers.json import DjangoJSONEncoder
 from django.utils.safestring import mark_safe
 
-from docutils.core import publish_parts
-from docutils.writers.html4css1 import Writer, HTMLTranslator
-
 import sentry_sdk
 
 from swh.web.common.origin_save import get_savable_visit_types
+from swh.web.common.utils import rst_to_html
 
 register = template.Library()
 
 
-class NoHeaderHTMLTranslator(HTMLTranslator):
-    """
-    Docutils translator subclass to customize the generation of HTML
-    from reST-formatted docstrings
-    """
-    def __init__(self, document):
-        super().__init__(document)
-        self.body_prefix = []
-        self.body_suffix = []
-
-    def visit_bullet_list(self, node):
-        self.context.append((self.compact_simple, self.compact_p))
-        self.compact_p = None
-        self.compact_simple = self.is_compactable(node)
-        self.body.append(self.starttag(node, 'ul', CLASS='docstring'))
-
-
-DOCSTRING_WRITER = Writer()
-DOCSTRING_WRITER.translator_class = NoHeaderHTMLTranslator
-
-
 @register.filter
 def safe_docstring_display(docstring):
     """
     Utility function to htmlize reST-formatted documentation in browsable
     api.
     """
-    docstring = cleandoc(docstring)
-    return publish_parts(docstring, writer=DOCSTRING_WRITER)['html_body']
+    return rst_to_html(cleandoc(docstring))
 
 
 @register.filter
diff --git a/swh/web/common/utils.py b/swh/web/common/utils.py
--- a/swh/web/common/utils.py
+++ b/swh/web/common/utils.py
@@ -3,8 +3,6 @@
 # License: GNU Affero General Public License version 3, or any later version
 # See top-level LICENSE file for more information
 
-import docutils.parsers.rst
-import docutils.utils
 import re
 
 from datetime import datetime, timezone
@@ -13,6 +11,12 @@
 
 from typing import Optional, Dict, Any
 
+import docutils.parsers.rst
+import docutils.utils
+
+from docutils.core import publish_parts
+from docutils.writers.html5_polyglot import Writer, HTMLTranslator
+
 from django.urls import reverse as django_reverse
 from django.http import QueryDict, HttpRequest
 
@@ -349,6 +353,7 @@
     Helper class to enforce CSRF validation on a DRF view
     when a user is not authenticated.
     """
+
     def authenticate(self, request):
         user = getattr(request._request, 'user', None)
         self.enforce_csrf(request)
@@ -424,12 +429,12 @@
         not be parsed.
     """
     pids_by_type = {
-            CONTENT: [],
-            DIRECTORY: [],
-            REVISION: [],
-            RELEASE: [],
-            SNAPSHOT: []
-            }
+        CONTENT: [],
+        DIRECTORY: [],
+        REVISION: [],
+        RELEASE: [],
+        SNAPSHOT: []
+    }
 
     for pid in persistent_ids:
         obj_id = pid.object_id
@@ -437,3 +442,38 @@
         pids_by_type[obj_type].append(hash_to_bytes(obj_id))
 
     return pids_by_type
+
+
+class _NoHeaderHTMLTranslator(HTMLTranslator):
+    """
+    Docutils translator subclass to customize the generation of HTML
+    from reST-formatted docstrings
+    """
+
+    def __init__(self, document):
+        super().__init__(document)
+        self.body_prefix = []
+        self.body_suffix = []
+
+
+_HTML_WRITER = Writer()
+_HTML_WRITER.translator_class = _NoHeaderHTMLTranslator
+
+
+def rst_to_html(rst: str) -> str:
+    """
+    Convert reStructuredText document into HTML.
+
+    Args:
+        rst: A string containing a reStructuredText document
+
+    Returns:
+        Body content of the produced HTML conversion.
+
+    """
+    settings = {
+        'initial_header_level': 2,
+    }
+    pp = publish_parts(rst, writer=_HTML_WRITER,
+                       settings_overrides=settings)
+    return f'<div class="swh-rst">{pp["html_body"]}</div>'
diff --git a/swh/web/templates/includes/readme-display.html b/swh/web/templates/includes/readme-display.html
--- a/swh/web/templates/includes/readme-display.html
+++ b/swh/web/templates/includes/readme-display.html
@@ -13,25 +13,25 @@
     <div class="card-header bg-gray-light">
       <h4>{{ readme_name }}</h4>
     </div>
-    <div id="readme" class="card-body">
+    <div class="swh-readme card-body">
     </div>
   </div>
 
   {% if readme_html %}
     <script>
-      $('#readme').html(swh.webapp.filterXSS({{ readme_html|jsonify }}));
+      $('.swh-readme').html(swh.webapp.filterXSS({{ readme_html|jsonify }}));
     </script>
   {% elif readme_name.lower == 'readme' or readme_name.lower == 'readme.txt' %}
     <script>
-      swh.webapp.renderTxt('#readme', '{{ readme_url }}');
+      swh.webapp.renderTxt('.swh-readme', '{{ readme_url }}');
     </script>
   {% elif readme_name.lower == 'readme.org' %}
     <script>
-      swh.webapp.renderOrg('#readme', '{{ readme_url }}');
+      swh.webapp.renderOrg('.swh-readme', '{{ readme_url }}');
     </script>
   {% else %}
     <script>
-      swh.webapp.renderMarkdown('#readme', '{{ readme_url }}');
+      swh.webapp.renderMarkdown('.swh-readme', '{{ readme_url }}');
     </script>
   {% endif %}
 
diff --git a/swh/web/tests/common/test_templatetags.py b/swh/web/tests/common/test_templatetags.py
--- a/swh/web/tests/common/test_templatetags.py
+++ b/swh/web/tests/common/test_templatetags.py
@@ -3,8 +3,6 @@
 # License: GNU Affero General Public License version 3, or any later version
 # See top-level LICENSE file for more information
 
-import pytest
-
 from swh.web.common.swh_templatetags import (
     urlize_links_and_mails, urlize_header_links, safe_docstring_display
 )
@@ -38,9 +36,6 @@
     assert urlize_header_links(content) == expected_content
 
 
-# remove deprecation warnings related to docutils
-@pytest.mark.filterwarnings(
-    'ignore:.*U.*mode is deprecated:DeprecationWarning')
 def test_safe_docstring_display():
     # update api link with html links content with links
     docstring = (
@@ -52,13 +47,15 @@
     )
 
     expected_docstring = (
+        '<div class="swh-rst">'
         '<p>This is my list header:</p>\n'
-        '<ul class="docstring">\n'
-        '<li>Here is item 1, with a continuation\n'
-        'line right here</li>\n'
-        '<li>Here is item 2</li>\n'
+        '<ul class="simple">\n'
+        '<li><p>Here is item 1, with a continuation\n'
+        'line right here</p></li>\n'
+        '<li><p>Here is item 2</p></li>\n'
         '</ul>\n'
         '<p>Here is something that is not part of the list</p>\n'
+        '</div>'
     )
 
     assert safe_docstring_display(docstring) == expected_docstring
diff --git a/swh/web/tests/common/test_utils.py b/swh/web/tests/common/test_utils.py
--- a/swh/web/tests/common/test_utils.py
+++ b/swh/web/tests/common/test_utils.py
@@ -112,3 +112,44 @@
     with pytest.raises(BadInputExc) as e:
         utils.get_swh_persistent_id(swh_object_type, 'not a valid id')
     assert e.match('Invalid object')
+
+
+def test_rst_to_html():
+    rst = (
+        'Section\n'
+        '=======\n\n'
+        '**Some strong text**\n\n'
+        'Subsection\n'
+        '----------\n\n'
+        '* This is a bulleted list.\n'
+        '* It has two items, the second\n'
+        '  item uses two lines.\n'
+        '\n'
+        '1. This is a numbered list.\n'
+        '2. It has two items too.\n'
+        '\n'
+        '#. This is a numbered list.\n'
+        '#. It has two items too.\n'
+    )
+
+    expected_html = (
+        '<div class="swh-rst"><h1 class="title">Section</h1>\n'
+        '<p><strong>Some strong text</strong></p>\n'
+        '<div class="section" id="subsection">\n'
+        '<h2>Subsection</h2>\n'
+        '<ul class="simple">\n'
+        '<li><p>This is a bulleted list.</p></li>\n'
+        '<li><p>It has two items, the second\n'
+        'item uses two lines.</p></li>\n'
+        '</ul>\n'
+        '<ol class="arabic simple">\n'
+        '<li><p>This is a numbered list.</p></li>\n'
+        '<li><p>It has two items too.</p></li>\n'
+        '<li><p>This is a numbered list.</p></li>\n'
+        '<li><p>It has two items too.</p></li>\n'
+        '</ol>\n'
+        '</div>\n'
+        '</div>'
+    )
+
+    assert utils.rst_to_html(rst) == expected_html