diff --git a/common/markdown.py b/common/markdown.py
new file mode 100644
index 0000000000..f0f056d963
--- /dev/null
+++ b/common/markdown.py
@@ -0,0 +1,45 @@
+HTML_REPLACEMENTS = [
+  (r'&', r'&'),
+  (r'"', r'"'),
+]
+
+def parse_markdown(text: str, tab_length: int = 2) -> str:
+  lines = text.split("\n")
+  output: list[str] = []
+  list_level = 0
+
+  def end_outstanding_lists(level: int, end_level: int) -> int:
+    while level > end_level:
+      level -= 1
+      output.append("</ul>")
+      if level > 0:
+        output.append("</li>")
+    return end_level
+
+  for i, line in enumerate(lines):
+    if i + 1 < len(lines) and lines[i + 1].startswith("==="):  # heading
+      output.append(f"<h1>{line}</h1>")
+    elif line.startswith("==="):
+      pass
+    elif line.lstrip().startswith("* "):  # list
+      line_level = 1 + line.count(" " * tab_length, 0, line.index("*"))
+      if list_level >= line_level:
+        list_level = end_outstanding_lists(list_level, line_level)
+      else:
+        list_level += 1
+        if list_level > 1:
+          output[-1] = output[-1].replace("</li>", "")
+        output.append("<ul>")
+      output.append(f"<li>{line.replace('*', '', 1).lstrip()}</li>")
+    else:
+      list_level = end_outstanding_lists(list_level, 0)
+      if len(line) > 0:
+        output.append(line)
+
+  end_outstanding_lists(list_level, 0)
+  output_str = "\n".join(output) + "\n"
+
+  for (fr, to) in HTML_REPLACEMENTS:
+    output_str = output_str.replace(fr, to)
+
+  return output_str
diff --git a/common/tests/test_markdown.py b/common/tests/test_markdown.py
new file mode 100644
index 0000000000..d3c7e02c69
--- /dev/null
+++ b/common/tests/test_markdown.py
@@ -0,0 +1,15 @@
+import os
+
+from openpilot.common.basedir import BASEDIR
+from openpilot.common.markdown import parse_markdown
+
+
+class TestMarkdown:
+  def test_all_release_notes(self):
+    with open(os.path.join(BASEDIR, "RELEASES.md")) as f:
+      release_notes = f.read().split("\n\n")
+      assert len(release_notes) > 10
+
+      for rn in release_notes:
+        md = parse_markdown(rn)
+        assert len(md) > 0
diff --git a/poetry.lock b/poetry.lock
index 062d14fdcb..7023c19f1a 100644
--- a/poetry.lock
+++ b/poetry.lock
@@ -1,4 +1,4 @@
-# This file is automatically @generated by Poetry 1.7.0 and should not be changed by hand.
+# This file is automatically @generated by Poetry 1.8.3 and should not be changed by hand.
 
 [[package]]
 name = "aiohttp"
@@ -8064,4 +8064,4 @@ test = ["big-O", "jaraco.functools", "jaraco.itertools", "jaraco.test", "more-it
 [metadata]
 lock-version = "2.0"
 python-versions = "~3.11"
-content-hash = "804b6450fde7f96b7bd127d8598d24645bf048aa094c625c3c0e43a3460fcf5e"
+content-hash = "e2473a1aa0d28b72f8a0816bb3335a2e48e526e39f038387194948965fffd72e"
diff --git a/pyproject.toml b/pyproject.toml
index 8642eff4cd..34f1e97ab2 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -121,7 +121,6 @@ casadi = "*"
 future-fstrings = "*"
 
 # these should be removed
-markdown-it-py = "*"
 timezonefinder = "*"
 setproctitle = "*"
 
diff --git a/system/updated/updated.py b/system/updated/updated.py
index c1af41bd92..d43f439af5 100755
--- a/system/updated/updated.py
+++ b/system/updated/updated.py
@@ -11,11 +11,11 @@ import time
 import threading
 from collections import defaultdict
 from pathlib import Path
-from markdown_it import MarkdownIt
 
 from openpilot.common.basedir import BASEDIR
 from openpilot.common.params import Params
 from openpilot.common.time import system_time_valid
+from openpilot.common.markdown import parse_markdown
 from openpilot.common.swaglog import cloudlog
 from openpilot.selfdrive.controls.lib.alertmanager import set_offroad_alert
 from openpilot.system.hardware import AGNOS, HARDWARE
@@ -89,7 +89,7 @@ def parse_release_notes(basedir: str) -> bytes:
     with open(os.path.join(basedir, "RELEASES.md"), "rb") as f:
       r = f.read().split(b'\n\n', 1)[0]  # Slice latest release notes
     try:
-      return bytes(MarkdownIt().render(r.decode("utf-8")), encoding="utf-8")
+      return bytes(parse_markdown(r.decode("utf-8")), encoding="utf-8")
     except Exception:
       return r + b"\n"
   except FileNotFoundError: