diff --git a/.github/join_profiles.py b/.github/join_profiles.py
new file mode 100644
index 00000000..09e4c13e
--- /dev/null
+++ b/.github/join_profiles.py
@@ -0,0 +1,466 @@
+#!/usr/bin/env python3
+"""Build root-level profile artifacts from the repo's split profile sources.
+
+Input layout:
+    <repo>/profile_sources/
+      <repo-id>/
+        <vendor-name>/
+          metadata.json
+          vendor.idx
+          vendor/
+          printer_model/
+          printer/
+          print/
+          filament/
+          physical_printer/
+          presets/
+          obsolete_presets/
+
+Output layout:
+    <repo>/
+      manifest.json
+      repos/
+        <repo-id>/
+          vendor_indices.zip
+          <vendor-name>/
+            <version>.ini
+"""
+
+from __future__ import annotations
+
+import argparse
+import io
+import json
+import sys
+import urllib.error
+import urllib.request
+import zipfile
+from collections import defaultdict
+from pathlib import Path
+from typing import Iterable
+from urllib.parse import quote, urljoin
+
+
+REPO_URLS = (
+    "https://preset-repo-api.prusa3d.com/v1/repos",
+    "https://raw.githubusercontent.com/Dark98/SliceBeam/refs/heads/master/.profiledumpsrepo/manifest.json",
+)
+
+USER_AGENT = "SliceBeamProfileDump/1.0"
+INVALID_FILE_CHARS = '<>:"/\\|?*'
+ASSET_KEYS = {"thumbnail", "bed_model", "bed_texture"}
+REPO_ROOT = Path(__file__).resolve().parent.parent
+PROFILE_SOURCES_DIR = REPO_ROOT / "profile_sources"
+BACKEND_STATIC_DIR = REPO_ROOT
+
+
+def fetch_bytes(url: str) -> bytes:
+    request = urllib.request.Request(url, headers={"User-Agent": USER_AGENT})
+    with urllib.request.urlopen(request) as response:
+        return response.read()
+
+
+def fetch_json(url: str):
+    return json.loads(fetch_bytes(url).decode("utf-8"))
+
+
+def parse_vendor_version(index_bytes: bytes) -> str:
+    for raw_line in index_bytes.decode("utf-8").splitlines():
+        line = raw_line.strip()
+        if not line or " = " in line:
+            continue
+        space_index = line.find(" ")
+        if space_index == -1:
+            raise ValueError(f"Malformed vendor index line: {line!r}")
+        return line[:space_index]
+    raise ValueError("Could not determine vendor version from index file")
+
+
+def sanitize_file_name(name: str) -> str:
+    sanitized = "".join("_" if ch in INVALID_FILE_CHARS else ch for ch in name).strip()
+    sanitized = sanitized.rstrip(". ")
+    return sanitized or "unnamed"
+
+
+def safe_asset_relative_path(path_str: str) -> Path:
+    normalized = path_str.replace("\\", "/").strip().lstrip("/")
+    if not normalized:
+        raise ValueError("Empty asset path")
+    candidate = Path(*[part for part in normalized.split("/") if part not in ("", ".")])
+    if any(part == ".." for part in candidate.parts):
+        raise ValueError(f"Unsafe asset path: {path_str}")
+    return candidate
+
+
+def parse_sections(ini_text: str) -> dict[str, list[tuple[str, list[str]]]]:
+    grouped: dict[str, list[tuple[str, list[str]]]] = defaultdict(list)
+    for section_type, section_name, lines in parse_sections_ordered(ini_text):
+        grouped[section_type].append((section_name, lines))
+    return dict(grouped)
+
+
+def parse_sections_ordered(ini_text: str) -> list[tuple[str, str, list[str]]]:
+    sections: dict[str, list[tuple[str, list[str]]]] = defaultdict(list)
+    current_type: str | None = None
+    current_name: str | None = None
+    current_lines: list[str] = []
+    ordered: list[tuple[str, str, list[str]]] = []
+
+    def flush() -> None:
+        nonlocal current_type, current_name, current_lines
+        if current_type is None:
+            return
+        name = current_name or current_type
+        sections[current_type].append((name, current_lines.copy()))
+        ordered.append((current_type, name, current_lines.copy()))
+        current_type = None
+        current_name = None
+        current_lines = []
+
+    for line in ini_text.splitlines():
+        if line.startswith("[") and line.endswith("]"):
+            flush()
+            section = line[1:-1]
+            if ":" in section:
+                section_type, section_name = section.split(":", 1)
+            else:
+                section_type, section_name = section, section
+            current_type = section_type
+            current_name = section_name
+            current_lines = [line]
+        elif current_type is not None:
+            current_lines.append(line)
+
+    flush()
+    return ordered
+
+
+def iter_referenced_assets(ordered_sections: list[tuple[str, str, list[str]]]) -> list[str]:
+    referenced: list[str] = []
+    seen: set[str] = set()
+    for section_type, _, lines in ordered_sections:
+        if section_type != "printer_model":
+            continue
+        for line in lines[1:]:
+            stripped = line.strip()
+            if not stripped or stripped.startswith(";") or stripped.startswith("#") or " = " not in stripped:
+                continue
+            key, value = stripped.split(" = ", 1)
+            if key not in ASSET_KEYS:
+                continue
+            value = value.strip()
+            if not value or value in seen:
+                continue
+            safe_asset_relative_path(value)
+            seen.add(value)
+            referenced.append(value)
+    return referenced
+
+
+def fetch_optional_asset(base_url: str, asset_path: str) -> bytes | None:
+    safe_rel_path = safe_asset_relative_path(asset_path).as_posix()
+    encoded_path = "/".join(quote(part) for part in safe_rel_path.split("/"))
+    asset_url = urljoin(base_url.rstrip("/") + "/", encoded_path)
+    try:
+        return fetch_bytes(asset_url)
+    except urllib.error.HTTPError as exc:
+        if exc.code == 404:
+            print(f"    [warn] missing asset {asset_path} at {asset_url}")
+            return None
+        raise
+
+
+def write_assets(assets_root: Path, assets: dict[str, bytes]) -> list[str]:
+    written: list[str] = []
+    for rel_path_str, data in sorted(assets.items()):
+        rel_path = safe_asset_relative_path(rel_path_str)
+        target = assets_root / rel_path
+        target.parent.mkdir(parents=True, exist_ok=True)
+        target.write_bytes(data)
+        written.append(rel_path.as_posix())
+    return written
+
+
+def collect_split_source_assets(vendor_dir: Path) -> dict[str, bytes]:
+    assets_dir = vendor_dir / "assets"
+    if not assets_dir.exists():
+        return {}
+
+    assets: dict[str, bytes] = {}
+    for path in sorted(p for p in assets_dir.rglob("*") if p.is_file()):
+        rel_path = path.relative_to(assets_dir).as_posix()
+        safe_asset_relative_path(rel_path)
+        assets[rel_path] = path.read_bytes()
+    return assets
+
+
+def write_sections(split_dir: Path, ini_text: str) -> dict[str, int]:
+    counts: dict[str, int] = {}
+    for section_type, _, _ in parse_sections_ordered(ini_text):
+        counts[section_type] = counts.get(section_type, 0) + 1
+    _write_split_sections(split_dir, parse_sections_ordered(ini_text))
+    return counts
+
+
+def _write_split_sections(split_dir: Path, ordered_sections: list[tuple[str, str, list[str]]]) -> list[dict[str, str]]:
+    file_usage: dict[tuple[str, str], int] = defaultdict(int)
+    section_order: list[dict[str, str]] = []
+    for section_type, section_name, lines in ordered_sections:
+        target_dir = split_dir / sanitize_file_name(section_type)
+        target_dir.mkdir(parents=True, exist_ok=True)
+
+        file_usage[(section_type, section_name)] += 1
+        suffix = file_usage[(section_type, section_name)]
+        file_name = sanitize_file_name(section_name)
+        if not file_name.lower().endswith(".ini"):
+            file_name = f"{file_name}.ini"
+        if suffix > 1:
+            stem = Path(file_name).stem
+            file_name = f"{stem}__{suffix}{Path(file_name).suffix}"
+
+        relative_path = Path(sanitize_file_name(section_type)) / file_name
+        (split_dir / relative_path).write_text("\n".join(lines) + "\n", encoding="utf-8")
+        section_order.append(
+            {
+                "type": section_type,
+                "name": section_name,
+                "path": relative_path.as_posix(),
+            }
+        )
+    return section_order
+
+
+def iter_filtered_repos(repo_filter: set[str] | None) -> list[dict]:
+    all_repos: list[dict] = []
+    for manifest_url in REPO_URLS:
+        entries = fetch_json(manifest_url)
+        for entry in entries:
+            if not entry["id"].endswith("-fff"):
+                continue
+            if repo_filter and entry["id"] not in repo_filter:
+                continue
+            all_repos.append(entry)
+    return all_repos
+
+
+def iter_vendor_entries(index_zip_bytes: bytes) -> Iterable[tuple[str, bytes]]:
+    with zipfile.ZipFile(io.BytesIO(index_zip_bytes)) as zf:
+        for name in sorted(zf.namelist()):
+            if name.endswith("/"):
+                continue
+            with zf.open(name) as fp:
+                yield name, fp.read()
+
+
+def dump_vendor(repo: dict, vendor_index_name: str, vendor_index_bytes: bytes, output_root: Path) -> tuple[dict, dict[str, bytes]]:
+    vendor_name = Path(vendor_index_name).stem
+    version = parse_vendor_version(vendor_index_bytes)
+    ini_url = repo["url"].rstrip("/") + f"/{vendor_name}/{version}.ini"
+    ini_bytes = fetch_bytes(ini_url)
+    ini_text = ini_bytes.decode("utf-8")
+    base_url = repo["url"].rstrip("/") + f"/{vendor_name}"
+    ordered_sections = parse_sections_ordered(ini_text)
+    referenced_assets = iter_referenced_assets(ordered_sections)
+    assets: dict[str, bytes] = {}
+    for asset_path in referenced_assets:
+        asset_bytes = fetch_optional_asset(base_url, asset_path)
+        if asset_bytes is not None:
+            assets[asset_path] = asset_bytes
+
+    vendor_dir = output_root / repo["id"] / sanitize_file_name(vendor_name)
+    raw_dir = vendor_dir / "raw"
+    split_dir = vendor_dir / "split"
+    assets_dir = vendor_dir / "assets"
+    raw_dir.mkdir(parents=True, exist_ok=True)
+    split_dir.mkdir(parents=True, exist_ok=True)
+    if assets:
+        assets_dir.mkdir(parents=True, exist_ok=True)
+
+    raw_path = raw_dir / f"{sanitize_file_name(version)}.ini"
+    raw_path.write_bytes(ini_bytes)
+    section_counts = write_sections(split_dir, ini_text)
+    asset_paths = write_assets(assets_dir, assets) if assets else []
+
+    metadata = {
+        "repo_id": repo["id"],
+        "repo_name": repo.get("name"),
+        "vendor": vendor_name,
+        "version": version,
+        "ini_url": ini_url,
+        "index_url": repo["index_url"],
+        "section_counts": section_counts,
+        "asset_paths": asset_paths,
+    }
+    (vendor_dir / "metadata.json").write_text(json.dumps(metadata, indent=2), encoding="utf-8")
+    return metadata, assets
+
+
+def write_split_source_vendor(
+    split_root: Path,
+    repo: dict,
+    vendor_index_name: str,
+    vendor_index_bytes: bytes,
+    version: str,
+    ini_text: str,
+    assets: dict[str, bytes] | None = None,
+) -> None:
+    vendor_name = Path(vendor_index_name).stem
+    vendor_dir = split_root / sanitize_file_name(repo["id"]) / sanitize_file_name(vendor_name)
+    vendor_dir.mkdir(parents=True, exist_ok=True)
+
+    ordered_sections = parse_sections_ordered(ini_text)
+    section_order = _write_split_sections(vendor_dir, ordered_sections)
+    asset_paths = write_assets(vendor_dir / "assets", assets or {}) if assets else []
+    (vendor_dir / "vendor.idx").write_bytes(vendor_index_bytes)
+    metadata = {
+        "repo": {
+            "id": repo["id"],
+            "name": repo.get("name", repo["id"]),
+            "description": repo.get("description", ""),
+            "visibility": repo.get("visibility", ""),
+        },
+        "vendor": vendor_name,
+        "version": version,
+        "index_name": vendor_index_name,
+        "section_order": section_order,
+        "asset_paths": asset_paths,
+    }
+    (vendor_dir / "metadata.json").write_text(json.dumps(metadata, indent=2), encoding="utf-8")
+
+
+def write_backend_static_repo(static_root: Path, repos: list[dict], vendor_artifacts: dict[str, list[dict]]) -> None:
+    static_root.mkdir(parents=True, exist_ok=True)
+    repos_root = static_root / "repos"
+    repos_root.mkdir(parents=True, exist_ok=True)
+
+    manifest_entries = []
+    for repo in repos:
+        repo_id = repo["id"]
+        repo_root = repos_root / sanitize_file_name(repo_id)
+        repo_root.mkdir(parents=True, exist_ok=True)
+
+        artifacts = vendor_artifacts.get(repo_id, [])
+        for artifact in artifacts:
+            vendor_root = repo_root / sanitize_file_name(artifact["vendor_name"])
+            vendor_root.mkdir(parents=True, exist_ok=True)
+            (vendor_root / f"{sanitize_file_name(artifact['version'])}.ini").write_bytes(artifact["ini_bytes"])
+            for asset_rel_path, asset_bytes in sorted(artifact.get("assets", {}).items()):
+                asset_target = vendor_root / safe_asset_relative_path(asset_rel_path)
+                asset_target.parent.mkdir(parents=True, exist_ok=True)
+                asset_target.write_bytes(asset_bytes)
+
+        with zipfile.ZipFile(repo_root / "vendor_indices.zip", "w", compression=zipfile.ZIP_DEFLATED) as zf:
+            for artifact in artifacts:
+                zf.writestr(artifact["index_name"], artifact["index_bytes"])
+
+        manifest_entries.append(
+            {
+                "name": repo.get("name", repo_id),
+                "description": repo.get("description", ""),
+                "visibility": repo.get("visibility", ""),
+                "id": repo_id,
+                "url": f"./repos/{repo_id}",
+                "index_url": f"./repos/{repo_id}/vendor_indices.zip",
+            }
+        )
+
+    (static_root / "manifest.json").write_text(json.dumps(manifest_entries, indent=2), encoding="utf-8")
+
+
+def build_backend_static_from_split_source(split_root: Path, static_root: Path) -> None:
+    repos_map: dict[str, dict] = {}
+    vendor_artifacts: dict[str, list[dict]] = defaultdict(list)
+
+    for repo_dir in sorted(path for path in split_root.iterdir() if path.is_dir()):
+        for vendor_dir in sorted(path for path in repo_dir.iterdir() if path.is_dir()):
+            metadata_path = vendor_dir / "metadata.json"
+            if not metadata_path.exists():
+                raise FileNotFoundError(f"Missing metadata.json in {vendor_dir}")
+
+            metadata = json.loads(metadata_path.read_text(encoding="utf-8"))
+            repo = metadata["repo"]
+            repo_id = repo["id"]
+            repos_map[repo_id] = {
+                "id": repo_id,
+                "name": repo.get("name", repo_id),
+                "description": repo.get("description", ""),
+                "visibility": repo.get("visibility", ""),
+            }
+
+            parts: list[str] = []
+            for section in metadata["section_order"]:
+                section_path = vendor_dir / Path(section["path"])
+                content = section_path.read_text(encoding="utf-8").rstrip()
+                parts.append(content)
+            ini_text = "\n\n".join(parts) + "\n"
+
+            vendor_artifacts[repo_id].append(
+                {
+                    "vendor_name": metadata["vendor"],
+                    "index_name": metadata["index_name"],
+                    "index_bytes": (vendor_dir / "vendor.idx").read_bytes(),
+                    "version": metadata["version"],
+                    "ini_bytes": ini_text.encode("utf-8"),
+                    "assets": collect_split_source_assets(vendor_dir),
+                }
+            )
+
+    repos = sorted(repos_map.values(), key=lambda item: item["id"])
+    write_backend_static_repo(static_root, repos, vendor_artifacts)
+
+
+def build_backend_static_repo_in_repo() -> None:
+    build_backend_static_from_split_source(PROFILE_SOURCES_DIR, BACKEND_STATIC_DIR)
+
+
+def split_backend_static_repo(static_root: Path, split_root: Path) -> None:
+    manifest_path = static_root / "manifest.json"
+    if not manifest_path.exists():
+        raise FileNotFoundError(f"Missing manifest.json in {static_root}")
+
+    repos = json.loads(manifest_path.read_text(encoding="utf-8"))
+    for repo in repos:
+        repo_id = repo["id"]
+        repo_dir = static_root / "repos" / sanitize_file_name(repo_id)
+        index_zip_path = repo_dir / "vendor_indices.zip"
+        if not index_zip_path.exists():
+            raise FileNotFoundError(f"Missing vendor_indices.zip for repo {repo_id}")
+
+        with zipfile.ZipFile(index_zip_path) as zf:
+            for index_name in sorted(zf.namelist()):
+                if index_name.endswith("/"):
+                    continue
+                vendor_name = Path(index_name).stem
+                index_bytes = zf.read(index_name)
+                version = parse_vendor_version(index_bytes)
+                ini_path = repo_dir / sanitize_file_name(vendor_name) / f"{sanitize_file_name(version)}.ini"
+                if not ini_path.exists():
+                    raise FileNotFoundError(f"Missing INI for {repo_id}/{vendor_name}/{version}")
+                ini_text = ini_path.read_text(encoding="utf-8")
+                ordered_sections = parse_sections_ordered(ini_text)
+                assets: dict[str, bytes] = {}
+                for asset_rel_path in iter_referenced_assets(ordered_sections):
+                    asset_path = repo_dir / sanitize_file_name(vendor_name) / safe_asset_relative_path(asset_rel_path)
+                    if asset_path.exists():
+                        assets[asset_rel_path] = asset_path.read_bytes()
+                write_split_source_vendor(split_root, repo, index_name, index_bytes, version, ini_text, assets)
+
+
+def main(argv: list[str]) -> int:
+    parser = argparse.ArgumentParser(description=__doc__)
+    args = parser.parse_args(argv)
+
+    try:
+        build_backend_static_repo_in_repo()
+        print(f"Backend static repo written to {BACKEND_STATIC_DIR}")
+        return 0
+    except urllib.error.URLError as exc:
+        print(f"Network error: {exc}", file=sys.stderr)
+        return 1
+    except Exception as exc:  # pragma: no cover
+        print(f"Failed: {exc}", file=sys.stderr)
+        return 1
+
+
+if __name__ == "__main__":
+    raise SystemExit(main(sys.argv[1:]))
diff --git a/.github/workflows/join-profiles.yml b/.github/workflows/join-profiles.yml
new file mode 100644
index 00000000..7b820e98
--- /dev/null
+++ b/.github/workflows/join-profiles.yml
@@ -0,0 +1,62 @@
+name: Join Profiles
+
+permissions:
+  contents: write
+
+on:
+  workflow_dispatch:
+  push:
+    paths:
+      - ".github/join_profiles.py"
+      - ".github/workflows/join-profiles.yml"
+      - "profile_sources/**"
+  pull_request:
+    paths:
+      - ".github/join_profiles.py"
+      - ".github/workflows/join-profiles.yml"
+      - "profile_sources/**"
+
+jobs:
+  join-profiles:
+    runs-on: ubuntu-latest
+    env:
+      PYTHON_VERSION: "3.12"
+    steps:
+      - name: Check out repository
+        uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
+
+      - name: Set up Python
+        uses: actions/setup-python@v5
+        with:
+          python-version: ${{ env.PYTHON_VERSION }}
+
+      - name: Join Profiles
+        run: |
+          echo "Joining Profiles"
+          echo "  Python: $(python --version)"
+          echo "  Script: $GITHUB_WORKSPACE/.github/join_profiles.py"
+          echo "  Output: $GITHUB_WORKSPACE"
+
+          rm -rf repos manifest.json
+
+          python "$GITHUB_WORKSPACE/.github/join_profiles.py"
+
+          echo "Profile joining finished"
+
+      - name: Commit generated profiles
+        if: github.event_name != 'pull_request'
+        run: |
+          git config user.name "github-actions[bot]"
+          git config user.email "41898282+github-actions[bot]@users.noreply.github.com"
+
+          git add manifest.json repos
+
+          if git diff --cached --quiet; then
+            echo "No generated profile changes to commit"
+            exit 0
+          fi
+
+          git commit -m "Update Profiles"
+          git push