diff --git a/requirements-dev.txt b/requirements-dev.txt
index 189fd5a7..2fad45f1 100644
--- a/requirements-dev.txt
+++ b/requirements-dev.txt
@@ -8,6 +8,7 @@ transifex-client==0.14.2
 selenium==3.141.0
 twine==3.4.1
 Pallets-Sphinx-Themes==1.2.3
+docutils==0.16
 Sphinx==3.5.3
 sphinx-issues==1.2.0
 sphinx-jinja==1.1.1
diff --git a/searx/engines/wordnik.py b/searx/engines/wordnik.py
new file mode 100644
index 00000000..3abe9efa
--- /dev/null
+++ b/searx/engines/wordnik.py
@@ -0,0 +1,77 @@
+# SPDX-License-Identifier: AGPL-3.0-or-later
+"""Wordnik (general)
+
+"""
+
+from lxml.html import fromstring
+from searx import logger
+from searx.utils import extract_text
+from searx.raise_for_httperror import raise_for_httperror
+
+logger = logger.getChild('Wordnik engine')
+
+# about
+about = {
+    "website": 'https://www.wordnik.com',
+    "wikidata_id": 'Q8034401',
+    "official_api_documentation": None,
+    "use_official_api": False,
+    "require_api_key": False,
+    "results": 'HTML',
+}
+
+categories = ['general']
+paging = False
+
+URL = 'https://www.wordnik.com'
+SEARCH_URL = URL + '/words/{query}'
+
+
+def request(query, params):
+    params['url'] = SEARCH_URL.format(query=query)
+    logger.debug(f"query_url --> {params['url']}")
+    return params
+
+
+def response(resp):
+    results = []
+
+    raise_for_httperror(resp)
+    dom = fromstring(resp.text)
+    word = extract_text(dom.xpath('//*[@id="headword"]/text()'))
+
+    definitions = []
+    for src in dom.xpath('//*[@id="define"]//h3[@class="source"]'):
+        src_text = extract_text(src).strip()
+        if src_text.startswith('from '):
+            src_text = src_text[5:]
+
+        src_defs = []
+        for def_item in src.xpath('following-sibling::ul[1]/li'):
+            def_abbr = extract_text(def_item.xpath('.//abbr')).strip()
+            def_text = extract_text(def_item).strip()
+            if def_abbr:
+                def_text = def_text[len(def_abbr):].strip()
+            src_defs.append((def_abbr, def_text))
+
+        definitions.append((src_text, src_defs))
+
+    if not definitions:
+        return results
+
+    infobox = ''
+    for src_text, src_defs in definitions:
+        infobox += f"<small>{src_text}</small>"
+        infobox += "<ul>"
+        for def_abbr, def_text in src_defs:
+            if def_abbr:
+                def_abbr += ": "
+            infobox += f"<li><i>{def_abbr}</i> {def_text}</li>"
+        infobox += "</ul>"
+
+    results.append({
+        'infobox': word,
+        'content': infobox,
+    })
+
+    return results
diff --git a/searx/settings.yml b/searx/settings.yml
index 81ef737d..8f2e50bc 100644
--- a/searx/settings.yml
+++ b/searx/settings.yml
@@ -1271,6 +1271,14 @@ engines:
     categories: videos
     disabled : True
 
+  - name: wordnik
+    engine: wordnik
+    shortcut: def
+    base_url: https://www.wordnik.com/
+    categories: general
+    timeout: 5.0
+    disabled: True
+
   - name: słownik języka polskiego
     engine: sjp
     shortcut: sjp
diff --git a/searx/webapp.py b/searx/webapp.py
index 4c09f0ac..072f140c 100755
--- a/searx/webapp.py
+++ b/searx/webapp.py
@@ -1133,11 +1133,14 @@ class ReverseProxyPathFix:
 
             base_url = urlparse(settings['server']['base_url'])
             self.script_name = base_url.path
+            if self.script_name.endswith('/'):
+                # remove trailing slash to avoid infinite redirect on the index
+                # see https://github.com/searx/searx/issues/2729
+                self.script_name = self.script_name[:-1]
             self.scheme = base_url.scheme
             self.server = base_url.netloc
 
     def __call__(self, environ, start_response):
-
         script_name = self.script_name or environ.get('HTTP_X_SCRIPT_NAME', '')
         if script_name:
             environ['SCRIPT_NAME'] = script_name