diff --git a/boost/src/chat.py b/boost/src/chat.py
index b3f252a..41d1521 100644
--- a/boost/src/chat.py
+++ b/boost/src/chat.py
@@ -49,6 +49,10 @@ def plain(self):
   def history(self):
     return self.tail.history()
 
+  def text(self):
+    # __str__ already does exactly this
+    return f"{self}"
+
   def __create_node(self, **kwargs):
     NodeType = self.chat_node_type
     return NodeType(**kwargs)
diff --git a/boost/src/custom_modules/discussurl.py b/boost/src/custom_modules/discussurl.py
new file mode 100644
index 0000000..9c6d50b
--- /dev/null
+++ b/boost/src/custom_modules/discussurl.py
@@ -0,0 +1,40 @@
+import re
+import requests
+
+url_regex = r"(?i)\b((?:https?://|www\d{0,3}[.]|[a-z0-9.\-]+[.][a-z]{2,4}/)(?:[^\s()<>]+|\(([^\s()<>]+|(\([^\s()<>]+\)))*\))+(?:\(([^\s()<>]+|(\([^\s()<>]+\)))*\)|[^\s`!()\[\]{};:'\".,<>?«»“”‘’]))"
+prompt = """
+<instruction>
+Your task is to fulfill the user's request by discussing provided content.
+</instruction>
+
+<content>
+{content}
+</content>
+
+<request>
+{request}
+</request>
+""".strip()
+
+ID_PREFIX = "discussurl"
+
+
+async def apply(chat, llm):
+  text = chat.text()
+  urls = re.findall(url_regex, text)
+
+  # No - URLs - proceed as usual
+  if len(urls) == 0:
+    return await llm.stream_final_completion()
+
+  # Yes - URLs - read them
+  content = ""
+  for url in urls:
+    await llm.emit_status(f"Reading {url[0]}...")
+    content += requests.get(url[0]).text
+
+  await llm.stream_final_completion(
+    prompt=prompt,
+    content=content,
+    request=chat.tail.content,
+  )