diff --git a/README.md b/README.md
index e8f1b55..faf0a45 100644
--- a/README.md
+++ b/README.md
@@ -129,6 +129,25 @@ All shift types are configurable through the in-app settings dialog.
    - Navigate to `http://localhost:5000`
    - Sign in with your Google account
 
+## Scripts
+
+### `scripts/extract_icons.py` — Extract icons from a PEP shift plan PDF
+
+Reads the Dienst-Legende on the last page of a POLYPOINT/PEP PDF, matches each icon to its shift code, and saves the icons as PNG files into `config/icons/`.
+
+**Requirements**
+```bash
+pip install pymupdf
+```
+
+**Usage**
+```bash
+python scripts/extract_icons.py "Mai Plan Definitiv.pdf"
+python scripts/extract_icons.py path/to/plan.pdf --out config/icons
+```
+
+The output directory defaults to `Server/config/icons`, which is the path served as `/icons` by the application.
+
 ## Documentation
 
 - [Authentication Setup Guide](authentication-setup.md) — Google OAuth configuration
diff --git a/scripts/extract_icons.py b/scripts/extract_icons.py
new file mode 100644
index 0000000..539d128
--- /dev/null
+++ b/scripts/extract_icons.py
@@ -0,0 +1,94 @@
+#!/usr/bin/env python3
+"""
+Extract shift-type icons from a POLYPOINT/PEP shift plan PDF.
+
+The script reads the Dienst-Legende on the last page of the PDF, matches each
+icon's position to the nearby shift code and name, then saves the icons as
+JPEG files into config/icons/.
+
+Usage:
+    python scripts/extract_icons.py "Mai Plan Definitiv.pdf"
+    python scripts/extract_icons.py path/to/plan.pdf --out config/icons
+
+Requirements:
+    pip install pymupdf
+"""
+
+import argparse
+import os
+import sys
+
+try:
+    import fitz
+except ImportError:
+    sys.exit("PyMuPDF is required: pip install pymupdf")
+
+
+def extract_icons(pdf_path: str, out_dir: str) -> None:
+    doc = fitz.open(pdf_path)
+    legend_page = doc[-1]
+
+    # Collect image bounding boxes: xref -> list of Rect
+    xref_to_rects: dict[int, list] = {}
+    for img in legend_page.get_images(full=True):
+        xref = img[0]
+        rects = legend_page.get_image_rects(xref)
+        xref_to_rects[xref] = list(rects)
+
+    # Collect text spans with positions
+    spans: list[dict] = []
+    for block in legend_page.get_text("dict")["blocks"]:
+        if block["type"] != 0:
+            continue
+        for line in block["lines"]:
+            for span in line["spans"]:
+                spans.append({"text": span["text"].strip(), "bbox": span["bbox"]})
+
+    def text_near(x: float, y: float, dx: float = 40, dy: float = 6) -> str:
+        """Return concatenated text of spans within (dx, dy) of (x, y)."""
+        matches = [
+            s["text"]
+            for s in spans
+            if abs(s["bbox"][0] - x) < dx and abs(s["bbox"][1] - y) < dy and s["text"]
+        ]
+        return " ".join(matches)
+
+    os.makedirs(out_dir, exist_ok=True)
+    saved = 0
+
+    for xref, rects in xref_to_rects.items():
+        img_info = doc.extract_image(xref)
+        w, h = img_info["width"], img_info["height"]
+
+        for rect in rects:
+            # Look for a shift code (numeric) just to the right of the icon
+            nearby = text_near(rect.x1, rect.y0, dx=50, dy=8)
+            parts = nearby.split()
+            code = next((p for p in parts if p.isdigit()), None)
+            if code is None:
+                continue
+
+            filename = f"{code}.png"
+            path = os.path.join(out_dir, filename)
+            fitz.Pixmap(img_info["image"]).save(path)
+            print(f"Saved {path} ({w}x{h} png)  —  {nearby}")
+            saved += 1
+            break  # one file per xref
+
+    print(f"\n{saved} icons saved to {out_dir}/")
+
+
+def main() -> None:
+    parser = argparse.ArgumentParser(description="Extract shift icons from a PEP PDF.")
+    parser.add_argument("pdf", help="Path to the shift plan PDF")
+    parser.add_argument("--out", default="Server/config/icons", help="Output directory (default: Server/config/icons)")
+    args = parser.parse_args()
+
+    if not os.path.isfile(args.pdf):
+        sys.exit(f"File not found: {args.pdf}")
+
+    extract_icons(args.pdf, args.out)
+
+
+if __name__ == "__main__":
+    main()