diff --git a/codesnap/output/markdown_exporter.py b/codesnap/output/markdown_exporter.py
new file mode 100644
index 0000000..582df30
--- /dev/null
+++ b/codesnap/output/markdown_exporter.py
@@ -0,0 +1,124 @@
+"""Markdown export module for CodeSnap."""
+
+from pathlib import Path
+from typing import Any, Optional
+
+from ..core.extractor import ExtractedCode
+
+
+class MarkdownExporter:
+    """Exports code summaries in Markdown format."""
+
+    def export(
+        self,
+        extracted_files: list[ExtractedCode],
+        file_paths: list[Path],
+        dependency_data: dict[str, Any],
+        complexity_data: dict[str, str],
+        output_path: Optional[Path] = None,
+    ) -> str:
+        """Export code summary to Markdown."""
+        lines: list[str] = []
+
+        lines.append("# CodeSnap Analysis Report\n")
+        lines.append(f"**Analyzed Directory:** {file_paths[0].parent if file_paths else 'N/A'}\n")
+        lines.append(f"**Files Analyzed:** {len(extracted_files)}\n")
+        lines.append(f"**Date:** {self._get_timestamp()}\n")
+
+        lines.append("\n## Language Breakdown\n")
+        language_counts: dict[str, int] = {}
+        for extracted in extracted_files:
+            language_counts[extracted.language] = language_counts.get(extracted.language, 0) + 1
+
+        for lang, count in sorted(language_counts.items(), key=lambda x: x[1], reverse=True):
+            lines.append(f"- **{lang}:** {count} files\n")
+
+        lines.append("\n## File Structure\n")
+        lines.append("```\n")
+        lines.append(self._generate_tree(file_paths))
+        lines.append("```\n")
+
+        lines.append("\n## Key Functions and Classes\n")
+        for extracted in extracted_files:
+            if extracted.functions or extracted.classes:
+                lines.append(f"\n### {extracted.file_path.name}\n")
+
+                if extracted.classes:
+                    lines.append("**Classes:**\n")
+                    for cls in extracted.classes:
+                        lines.append(f"- `{cls.name}`")
+                        if cls.base_classes:
+                            lines.append(f" (extends: {', '.join(cls.base_classes)})")
+                        lines.append("\n")
+                        if cls.methods:
+                            lines.append("  **Methods:**\n")
+                            for method in cls.methods:
+                                params = ", ".join(method.parameters) if method.parameters else ""
+                                lines.append(f"  - `{method.name}({params})`\n")
+
+                if extracted.functions:
+                    lines.append("**Functions:**\n")
+                    for func in extracted.functions:
+                        params = ", ".join(func.parameters) if func.parameters else ""
+                        lines.append(f"- `{func.name}({params})`\n")
+
+        lines.append("\n## Dependencies\n")
+        dependencies = dependency_data.get("dependencies", [])
+        if dependencies:
+            lines.append("### Import Relationships\n")
+            for dep in dependencies[:20]:
+                lines.append(f"- `{dep.source.name}` → `{dep.target.name}`\n")
+
+            if len(dependencies) > 20:
+                lines.append(f"\n... and {len(dependencies) - 20} more dependencies\n")
+        else:
+            lines.append("No dependencies detected.\n")
+
+        cycles = dependency_data.get("cycles", [])
+        if cycles:
+            lines.append("\n### Circular Dependencies\n")
+            for i, cycle in enumerate(cycles, 1):
+                cycle_names = " → ".join(p.name for p in cycle)
+                lines.append(f"**Cycle {i}:** {cycle_names}\n")
+
+        orphaned = dependency_data.get("orphaned", [])
+        if orphaned:
+            lines.append("\n### Orphaned Files (no dependencies)\n")
+            for f in orphaned:
+                lines.append(f"- {f.name}\n")
+
+        lines.append("\n## Complexity Metrics\n")
+        lines.append("| File | Complexity |\n")
+        lines.append("|------|------------|\n")
+        for extracted in extracted_files:
+            complexity = complexity_data.get(str(extracted.file_path), "unknown")
+            lines.append(f"| {extracted.file_path.name} | {complexity} |\n")
+
+        report = "\n".join(lines)
+
+        if output_path:
+            output_path.write_text(report, encoding="utf-8")
+
+        return report
+
+    def _generate_tree(self, file_paths: list[Path]) -> str:
+        """Generate a tree representation of the file structure."""
+        if not file_paths:
+            return ""
+
+        root = file_paths[0].parent
+        lines: list[str] = []
+
+        for path in sorted(file_paths):
+            relative = path.relative_to(root)
+            parts = list(relative.parts)
+            indent = "  " * (len(parts) - 1)
+            prefix = "└── " if parts[-1] == relative.parts[-1] else "├── "
+            lines.append(f"{indent}{prefix}{parts[-1]}\n")
+
+        return "".join(lines)
+
+    def _get_timestamp(self) -> str:
+        """Get current timestamp."""
+        from datetime import datetime
+        return datetime.utcnow().strftime("%Y-%m-%d %H:%M UTC")