diff --git a/Dockerfile b/Dockerfile
index d06e4c8..70ccd87 100644
--- a/Dockerfile
+++ b/Dockerfile
@@ -6,6 +6,7 @@ RUN apt-get update && apt-get install -y --no-install-recommends \
   texlive-xetex \
   texlive-lang-cyrillic \
   texlive-latex-extra \
+  texlive-fonts-recommended \
   fonts-liberation \
   cabextract \
   wget \
diff --git a/md2gost/handlers/__init__.py b/md2gost/handlers/__init__.py
index d7c466f..724d20a 100644
--- a/md2gost/handlers/__init__.py
+++ b/md2gost/handlers/__init__.py
@@ -1,5 +1,6 @@
-from md2gost.handlers.code import fence, hr
+from md2gost.handlers.code import fence
 from md2gost.handlers.heading import heading
+from md2gost.handlers.hr import hr
 from md2gost.handlers.image import image
 from md2gost.handlers.lists import bullet_list, ordered_list
 from md2gost.handlers.table import table
diff --git a/md2gost/handlers/code.py b/md2gost/handlers/code.py
index 73f879d..cbe4843 100644
--- a/md2gost/handlers/code.py
+++ b/md2gost/handlers/code.py
@@ -10,8 +10,5 @@ if TYPE_CHECKING:
 def fence(r: Renderer, tok: Token) -> str:
     lang = tok.info.strip()
     code = tok.content.rstrip()
-    return f"\\begin{{lstlisting}}[language={lang}]\n{code}\n\\end{{lstlisting}}"
-
-
-def hr(r: Renderer) -> str:
-    return r"\noindent\rule{\textwidth}{0.4pt}"
+    opts = f"[language={lang}]" if lang else ""
+    return f"\\begin{{lstlisting}}{opts}\n{code}\n\\end{{lstlisting}}"
diff --git a/md2gost/handlers/heading.py b/md2gost/handlers/heading.py
index ace926c..4c00178 100644
--- a/md2gost/handlers/heading.py
+++ b/md2gost/handlers/heading.py
@@ -5,6 +5,7 @@ from typing import TYPE_CHECKING
 from md2gost.handlers.inline import render_inline
 
 if TYPE_CHECKING:
+    from markdown_it.token import Token
     from md2gost.render import Renderer
 
 UNNUMBERED = {
@@ -24,21 +25,22 @@ NEWPAGE_BEFORE = {
 }
 
 
-def heading(r: Renderer) -> str:
+def heading(r: Renderer, tok: Token | None = None) -> str:
     tok = r.tokens[r.pos]
     level = int(tok.tag[1])
     inline_tok = r.tokens[r.pos + 1]
     text = render_inline(inline_tok)
 
-    r.current_heading = text.upper()
+    upper = text.upper()
+    r.current_heading = upper
     r.pos += 2
 
-    if text.upper() == TOC_HEADING:
+    if upper == TOC_HEADING:
         return ""
 
-    if text.upper() in UNNUMBERED:
+    if upper in UNNUMBERED:
         prefix = ""
-        if text.upper() in NEWPAGE_BEFORE:
+        if upper in NEWPAGE_BEFORE:
             prefix = "\\clearpage\n"
         return (
             f"{prefix}"
@@ -46,16 +48,10 @@ def heading(r: Renderer) -> str:
         )
 
     if level == 1:
-        r.section += 1
-        r.subsection = 0
-        r.subsubsection = 0
         return f"\\section{{{text}}}"
     elif level == 2:
-        r.subsection += 1
-        r.subsubsection = 0
         return f"\\subsection{{{text}}}"
     elif level == 3:
-        r.subsubsection += 1
         return f"\\subsubsection{{{text}}}"
     else:
         return f"\\textbf{{{text}}}"
diff --git a/md2gost/handlers/hr.py b/md2gost/handlers/hr.py
new file mode 100644
index 0000000..0807014
--- /dev/null
+++ b/md2gost/handlers/hr.py
@@ -0,0 +1,11 @@
+from __future__ import annotations
+
+from typing import TYPE_CHECKING
+
+if TYPE_CHECKING:
+    from markdown_it.token import Token
+    from md2gost.render import Renderer
+
+
+def hr(r: Renderer, tok: Token | None = None) -> str:
+    return r"\noindent\rule{\textwidth}{0.4pt}"
diff --git a/md2gost/handlers/inline.py b/md2gost/handlers/inline.py
index 1da0a96..734dd2a 100644
--- a/md2gost/handlers/inline.py
+++ b/md2gost/handlers/inline.py
@@ -5,10 +5,27 @@ from typing import TYPE_CHECKING
 if TYPE_CHECKING:
     from markdown_it.token import Token
 
+_LATEX_SPECIAL = str.maketrans({
+    "&": r"\&",
+    "%": r"\%",
+    "$": r"\$",
+    "#": r"\#",
+    "_": r"\_",
+    "{": r"\{",
+    "}": r"\}",
+    "~": r"\textasciitilde{}",
+    "^": r"\textasciicircum{}",
+    "\\": r"\textbackslash{}",
+})
+
+
+def escape_latex(text: str) -> str:
+    return text.translate(_LATEX_SPECIAL)
+
 
 def render_inline(tok: Token) -> str:
     if not tok.children:
-        return tok.content or ""
+        return escape_latex(tok.content or "")
     return render_inline_tokens(tok.children)
 
 
@@ -16,7 +33,7 @@ def render_inline_tokens(tokens: list[Token]) -> str:
     parts: list[str] = []
     for tok in tokens:
         if tok.type == "text":
-            parts.append(tok.content)
+            parts.append(escape_latex(tok.content))
         elif tok.type == "strong_open":
             parts.append(r"\textbf{")
         elif tok.type == "strong_close":
@@ -29,11 +46,11 @@ def render_inline_tokens(tokens: list[Token]) -> str:
             parts.append(f"\\verb|{tok.content}|")
         elif tok.type == "link_open":
             href = tok.attrGet("href") or ""
-            parts.append(f"\\href{{{href}}}{{")
+            parts.append(f"\\href{{{escape_latex(href)}}}{{")
         elif tok.type == "link_close":
             parts.append("}")
         elif tok.type in ("softbreak", "hardbreak"):
             parts.append("~\\\\")
         else:
-            parts.append(tok.content if tok.content else "")
+            parts.append(escape_latex(tok.content) if tok.content else "")
     return "".join(parts)
diff --git a/md2gost/handlers/lists.py b/md2gost/handlers/lists.py
index 4585a70..796eb0e 100644
--- a/md2gost/handlers/lists.py
+++ b/md2gost/handlers/lists.py
@@ -9,7 +9,7 @@ if TYPE_CHECKING:
     from md2gost.render import Renderer
 
 
-def bullet_list(r: Renderer) -> str:
+def bullet_list(r: Renderer, tok: Token | None = None) -> str:
     items = _collect_list_items(r)
     lines = []
     for item in items:
@@ -22,7 +22,7 @@ def bullet_list(r: Renderer) -> str:
     return f"\\begin{{itemize}}\n{body}\n\\end{{itemize}}"
 
 
-def ordered_list(r: Renderer) -> str:
+def ordered_list(r: Renderer, tok: Token | None = None) -> str:
     items = _collect_list_items(r)
     lines = []
     for item in items:
@@ -31,7 +31,8 @@ def ordered_list(r: Renderer) -> str:
             text = text.rstrip(".")
             lines.append(f"  \\item {text}.")
         else:
-            lines.append(f"  \\item {text.capitalize()}.")
+            first = text[0].upper() if text else text
+            lines.append(f"  \\item {first}{text[1:]}.")
     body = "\n".join(lines)
     return f"\\begin{{enumerate}}\n{body}\n\\end{{enumerate}}"
 
diff --git a/md2gost/handlers/table.py b/md2gost/handlers/table.py
index 0ce2111..f8e94db 100644
--- a/md2gost/handlers/table.py
+++ b/md2gost/handlers/table.py
@@ -8,7 +8,7 @@ if TYPE_CHECKING:
     from md2gost.render import Renderer
 
 
-def table(r: Renderer) -> str:
+def table(r: Renderer, tok: None = None) -> str:
     rows: list[list[str]] = []
     r.pos += 1
     while r.pos < len(r.tokens):
@@ -40,7 +40,10 @@ def table(r: Renderer) -> str:
     if not rows:
         return ""
 
-    cols = len(rows[0])
+    cols = max(len(row) for row in rows)
+    for row in rows:
+        row.extend([""] * (cols - len(row)))
+
     col_spec = "|".join(["l"] * cols)
     lines = [f"\\begin{{tabular}}{{|{col_spec}|}}", "\\hline"]
     for row in rows:
diff --git a/md2gost/handlers/text.py b/md2gost/handlers/text.py
index def5ce0..f3c8808 100644
--- a/md2gost/handlers/text.py
+++ b/md2gost/handlers/text.py
@@ -5,10 +5,11 @@ from typing import TYPE_CHECKING
 from md2gost.handlers.inline import render_inline
 
 if TYPE_CHECKING:
+    from markdown_it.token import Token
     from md2gost.render import Renderer
 
 
-def paragraph(r: Renderer) -> str:
+def paragraph(r: Renderer, tok: Token | None = None) -> str:
     inline_tok = r.tokens[r.pos + 1]
     text = render_inline(inline_tok)
     r.pos += 2
diff --git a/md2gost/latex/preamble.py b/md2gost/latex/preamble.py
index 69e1c4e..6de3ca9 100644
--- a/md2gost/latex/preamble.py
+++ b/md2gost/latex/preamble.py
@@ -93,4 +93,4 @@ PREAMBLE = r"""\documentclass[a4paper, 14pt]{extarticle}
 
 \begin{document}"""
 
-END_DOCUMENT = r"\end{document}"
+END_DOCUMENT = r"\end{document}" + "\n"
diff --git a/md2gost/latex/titlepage.py b/md2gost/latex/titlepage.py
index 10e0f14..0493727 100644
--- a/md2gost/latex/titlepage.py
+++ b/md2gost/latex/titlepage.py
@@ -1,31 +1,38 @@
 from __future__ import annotations
 
+from md2gost.handlers.inline import escape_latex
+
 
 def render_titlepage(fm: dict[str, str]) -> str:
-    faculty = fm.get("факультет", "")
-    department = fm.get("кафедра", "")
-    report_type = fm.get("тип", "")
-    title = fm.get("название", "")
-    student = fm.get("студент", "")
-    advisor = fm.get("преподаватель", "")
-    group = fm.get("группа", "")
-    year = fm.get("год", "2026")
+    required = {"название": "заголовок работы", "тип": "тип работы"}
+    for key, desc in required.items():
+        if key not in fm:
+            raise ValueError(f"front matter: отсутствует поле '{key}' ({desc})")
+
+    faculty = escape_latex(fm.get("факультет", ""))
+    department = escape_latex(fm.get("кафедра", ""))
+    report_type = escape_latex(fm.get("тип", ""))
+    title = escape_latex(fm.get("название", ""))
+    student = escape_latex(fm.get("студент", ""))
+    advisor = escape_latex(fm.get("преподаватель", ""))
+    group = escape_latex(fm.get("группа", ""))
+    year = escape_latex(fm.get("год", "2026"))
 
     ministry_lines = (
         "\\vspace*{-0.7ex}\n"
-        "\\hspace*{-6pt}\\centerline{Министерство науки и высшего образования Российской Федерации}\n"
+        f"\\hspace*{{-6pt}}\\centerline{{{fm.get('министерство', 'Министерство науки и высшего образования Российской Федерации')}}}\n"
         "\\vspace*{-0.7ex}\n"
-        "\\centerline{Федеральное государственное бюджетное образовательное учреждение }\n"
+        f"\\centerline{{{fm.get('вуз_тип', 'Федеральное государственное бюджетное образовательное учреждение ')}}}\n"
         "\\vspace*{-0.7ex}\n"
-        "\\centerline{высшего образования}\n"
+        f"\\centerline{{{fm.get('вуз_тип2', 'высшего образования')}}}\n"
         "\\vspace*{-0.7ex}\n"
-        "\\centerline{<<Московский государственный технический университет}\n"
+        f"\\centerline{{<<{fm.get('вуз', 'Московский государственный технический университет')}}}\n"
         "\\vspace*{-0.7ex}\n"
-        "\\centerline{имени Н.Э. Баумана>>}\n"
+        f"\\centerline{{имени Н.Э. Баумана>>}}\n"
         "\\vspace*{-0.7ex}\n"
-        "\\centerline{(национальный исследовательский университет)}\n"
+        f"\\centerline{{({fm.get('вуз_статус', 'национальный исследовательский университет')}}}\n"
         "\\vspace*{-0.7ex}\n"
-        "\\centerline{(МГТУ им. Н.Э. Баумана)}"
+        f"\\centerline{{({fm.get('вуз_аббревиатура', 'МГТУ им. Н.Э. Баумана')}}}"
     )
 
     titlepage = (
diff --git a/md2gost/main.py b/md2gost/main.py
index 968c193..36516db 100644
--- a/md2gost/main.py
+++ b/md2gost/main.py
@@ -9,6 +9,13 @@ from md2gost.parser import Parser
 from md2gost.render import Renderer
 
 
+def _print_latex_errors(log_path: Path) -> None:
+    if log_path.exists():
+        for line in log_path.read_text(errors="replace").splitlines():
+            if line.startswith("!"):
+                print(f"  {line}", file=sys.stderr)
+
+
 def run_xelatex(tex_path: Path, output_dir: Path) -> bool:
     result = subprocess.run(
         ["xelatex", "-interaction=nonstopmode",
@@ -16,10 +23,10 @@ def run_xelatex(tex_path: Path, output_dir: Path) -> bool:
         capture_output=True, text=True,
     )
     pdf_path = output_dir / tex_path.with_suffix(".pdf").name
-    return pdf_path.exists()
+    return pdf_path.exists() and result.returncode == 0
 
 
-def main():
+def main() -> None:
     cli = argparse.ArgumentParser(
         prog="md2gost",
         description="Markdown → LaTeX конвертер по ГОСТ Р 7.0.5-2008",
@@ -46,36 +53,28 @@ def main():
     renderer = Renderer(parser.tokens, parser.front_matter)
     latex = renderer.render()
 
-    # Собираем во временную директорию, копируем только .tex и .pdf
     with tempfile.TemporaryDirectory(prefix="md2gost-") as cache:
         cache_dir = Path(cache)
 
-        # Генерируем .tex в кэш
         tex_name = args.input.with_suffix(".tex").name
         tex_path = cache_dir / tex_name
         tex_path.write_text(latex, encoding="utf-8")
 
-        # Копируем .tex в выходную директорию
         final_tex = args.output / tex_name
         shutil.copy2(tex_path, final_tex)
         print(f"→ {final_tex}")
 
         if not args.no_pdf:
-            # Компилируем PDF в кэше (два прохода для TOC)
             if not run_xelatex(tex_path, cache_dir):
                 print("ошибка: xelatex не собрал PDF", file=sys.stderr)
-                log_path = cache_dir / "report.log"
-                if log_path.exists():
-                    for line in log_path.read_text(errors="replace").splitlines():
-                        if line.startswith("!"):
-                            print(f"  {line}", file=sys.stderr)
+                _print_latex_errors(cache_dir / "report.log")
                 sys.exit(1)
 
             if not run_xelatex(tex_path, cache_dir):
                 print("ошибка: xelatex (2-й проход) не собрал PDF", file=sys.stderr)
+                _print_latex_errors(cache_dir / "report.log")
                 sys.exit(1)
 
-            # Копируем .pdf в выходную директорию
             pdf_name = args.input.with_suffix(".pdf").name
             pdf_path = cache_dir / pdf_name
             final_pdf = args.output / pdf_name
diff --git a/md2gost/render.py b/md2gost/render.py
index 0d4e390..05764f9 100644
--- a/md2gost/render.py
+++ b/md2gost/render.py
@@ -16,18 +16,6 @@ from md2gost.latex import END_DOCUMENT, PREAMBLE, render_titlepage
 from markdown_it.token import Token
 
 
-def _fence(r, tok):
-    return fence(r, tok)
-
-
-def _image(r, tok):
-    return image(r, tok)
-
-
-def _inline(r, tok):
-    return render_inline(tok)
-
-
 _BLOCK = {
     "heading_open": heading,
     "paragraph_open": paragraph,
@@ -38,9 +26,9 @@ _BLOCK = {
 }
 
 _INLINE = {
-    "fence": _fence,
-    "image": _image,
-    "inline": _inline,
+    "fence": fence,
+    "image": image,
+    "inline": render_inline,
 }
 
 
@@ -49,9 +37,6 @@ class Renderer:
         self.tokens = tokens
         self.fm = front_matter
         self.pos = 0
-        self.section = 0
-        self.subsection = 0
-        self.subsubsection = 0
         self.current_heading = ""
 
     def render(self) -> str:
@@ -63,12 +48,13 @@ class Renderer:
             tt = tok.type
 
             if tt in _BLOCK:
-                result = _BLOCK[tt](self)
+                result = _BLOCK[tt](self, tok)
                 parts.append(result)
                 if self.current_heading == "СОДЕРЖАНИЕ":
                     parts.append("\\newpage\n\\tableofcontents\n\\newpage")
             elif tt in _INLINE:
-                parts.append(_INLINE[tt](self, tok))
+                result = _INLINE[tt](self, tok)
+                parts.append(result)
             else:
                 self.pos += 1
                 continue