cleanup tui, startup, card templtes, and more

2026-04-06 07:47:53 +10:00 · 2026-01-19 14:33:20 -08:00
parent bef67760db
commit 86e7973334
68 changed files with 1370 additions and 546 deletions
--- a/.claude/settings.json
+++ b/.claude/settings.json
@@ -0,0 +1,5 @@
+{
+  "enabledPlugins": {
+    "pyright-lsp@claude-plugins-official": true
+  }
+}
--- a/archivebox/cli/archivebox_server.py
+++ b/archivebox/cli/archivebox_server.py
@@ -30,10 +30,13 @@ def server(runserver_args: Iterable[str]=(SERVER_CONFIG.BIND_ADDR,),
    from archivebox.misc.checks import check_data_folder
    check_data_folder()

-    from django.core.management import call_command
-    from django.contrib.auth.models import User
-    
    from archivebox.config.common import SHELL_CONFIG
+
+    run_in_debug = SHELL_CONFIG.DEBUG or debug or reload
+    if debug or reload:
+        SHELL_CONFIG.DEBUG = True
+
+    from django.contrib.auth.models import User
    
    if not User.objects.filter(is_superuser=True).exclude(username='system').exists():
        print()
@@ -56,7 +59,8 @@ def server(runserver_args: Iterable[str]=(SERVER_CONFIG.BIND_ADDR,),
    except IndexError:
        pass

-    if SHELL_CONFIG.DEBUG:
+    if run_in_debug:
+        from django.core.management import call_command
        print('[green][+] Starting ArchiveBox webserver in DEBUG mode...[/green]')
        print(f'    [blink][green]>[/green][/blink] Starting ArchiveBox webserver on [deep_sky_blue4][link=http://{host}:{port}]http://{host}:{port}[/link][/deep_sky_blue4]')
        print(f'    [green]>[/green] Log in to ArchiveBox Admin UI on [deep_sky_blue3][link=http://{host}:{port}/admin]http://{host}:{port}/admin[/link][/deep_sky_blue3]')
--- a/archivebox/core/admin_archiveresults.py
+++ b/archivebox/core/admin_archiveresults.py
@@ -57,7 +57,7 @@ def render_archiveresults_list(archiveresults_qs, limit=50):

        # Build output link - use embed_path() which checks output_files first
        embed_path = result.embed_path() if hasattr(result, 'embed_path') else None
-        output_link = f'/archive/{result.snapshot.timestamp}/{embed_path}' if embed_path and result.status == 'succeeded' else f'/archive/{result.snapshot.timestamp}/'
+        output_link = f'/{result.snapshot.archive_path}/{embed_path}' if embed_path and result.status == 'succeeded' else f'/{result.snapshot.archive_path}/'

        # Get version - try cmd_version field
        version = result.cmd_version if result.cmd_version else '-'
@@ -83,8 +83,8 @@ def render_archiveresults_list(archiveresults_qs, limit=50):
                    {icon}
                </td>
                <td style="padding: 10px 12px; font-weight: 500; color: #334155;">
-                    <a href="{output_link}" target="_blank"
-                       style="color: #334155; text-decoration: none;"
+                        <a href="{output_link}" target="_blank"
+                           style="color: #334155; text-decoration: none;"
                       title="View output fullscreen"
                       onmouseover="this.style.color='#2563eb'; this.style.textDecoration='underline';"
                       onmouseout="this.style.color='#334155'; this.style.textDecoration='none';">
@@ -301,8 +301,8 @@ class ArchiveResultAdmin(BaseModelAdmin):
    )
    def snapshot_info(self, result):
        return format_html(
-            '<a href="/archive/{}/index.html"><b><code>[{}]</code></b> &nbsp; {} &nbsp; {}</a><br/>',
-            result.snapshot.timestamp,
+            '<a href="/{}/index.html"><b><code>[{}]</code></b> &nbsp; {} &nbsp; {}</a><br/>',
+            result.snapshot.archive_path,
            str(result.snapshot.id)[:8],
            result.snapshot.bookmarked_at.strftime('%Y-%m-%d %H:%M'),
            result.snapshot.url[:128],
@@ -336,8 +336,8 @@ class ArchiveResultAdmin(BaseModelAdmin):
        embed_path = result.embed_path() if hasattr(result, 'embed_path') else None
        output_path = embed_path if (result.status == 'succeeded' and embed_path) else 'index.html'
        return format_html(
-            '<a href="/archive/{}/{}" class="output-link">↗️</a><pre>{}</pre>',
-            result.snapshot.timestamp,
+            '<a href="/{}/{}" class="output-link">↗️</a><pre>{}</pre>',
+            result.snapshot.archive_path,
            output_path,
            result.output_str,
        )
@@ -348,7 +348,7 @@ class ArchiveResultAdmin(BaseModelAdmin):
            '<pre style="display: inline-block">{}</pre><br/>',
            result.output_str,
        )
-        output_html += format_html('<a href="/archive/{}/index.html#all">See result files ...</a><br/><pre><code>', str(result.snapshot.timestamp))
+        output_html += format_html('<a href="/{}/index.html#all">See result files ...</a><br/><pre><code>', str(result.snapshot.archive_path))
        embed_path = result.embed_path() if hasattr(result, 'embed_path') else ''
        path_from_embed = (snapshot_dir / (embed_path or ''))
        output_html += format_html('<i style="padding: 1px">{}</i><b style="padding-right: 20px">/</b><i>{}</i><br/><hr/>', str(snapshot_dir), str(embed_path))
--- a/archivebox/core/admin_snapshots.py
+++ b/archivebox/core/admin_snapshots.py
@@ -237,13 +237,13 @@ class SnapshotAdmin(SearchResultsAdminMixin, ConfigEditorMixin, BaseModelAdmin):
            '''
            <div style="display: flex; flex-wrap: wrap; gap: 12px; align-items: center;">
                <a class="btn" style="display: inline-flex; align-items: center; gap: 6px; padding: 10px 16px; background: #f8fafc; border: 1px solid #e2e8f0; border-radius: 8px; color: #334155; text-decoration: none; font-size: 14px; font-weight: 500; transition: all 0.15s;"
-                   href="/archive/{}"
+                   href="/{}"
                   onmouseover="this.style.background='#f1f5f9'; this.style.borderColor='#cbd5e1';"
                   onmouseout="this.style.background='#f8fafc'; this.style.borderColor='#e2e8f0';">
                    📄 Summary Page
                </a>
                <a class="btn" style="display: inline-flex; align-items: center; gap: 6px; padding: 10px 16px; background: #f8fafc; border: 1px solid #e2e8f0; border-radius: 8px; color: #334155; text-decoration: none; font-size: 14px; font-weight: 500; transition: all 0.15s;"
-                   href="/archive/{}/index.html#all"
+                   href="/{}/index.html#all"
                   onmouseover="this.style.background='#f1f5f9'; this.style.borderColor='#cbd5e1';"
                   onmouseout="this.style.background='#f8fafc'; this.style.borderColor='#e2e8f0';">
                    📁 Result Files
@@ -291,8 +291,8 @@ class SnapshotAdmin(SearchResultsAdminMixin, ConfigEditorMixin, BaseModelAdmin):
                <b>Tip:</b> Action buttons link to the list view with this snapshot pre-selected. Select it and use the action dropdown to execute.
            </p>
            ''',
-            obj.timestamp,
-            obj.timestamp,
+            obj.archive_path,
+            obj.archive_path,
            obj.url,
            obj.pk,
            obj.pk,
@@ -310,7 +310,7 @@ class SnapshotAdmin(SearchResultsAdminMixin, ConfigEditorMixin, BaseModelAdmin):
            '✅' if obj.is_archived else '❌',
            obj.num_outputs,
            self.size(obj) or '0kb',
-            f'/archive/{obj.timestamp}/favicon.ico',
+            f'/{obj.archive_path}/favicon.ico',
            obj.extension or '-',
        )

--- a/archivebox/core/middleware.py
+++ b/archivebox/core/middleware.py
@@ -1,6 +1,7 @@
 __package__ = 'archivebox.core'

 import ipaddress
+import re
 from django.utils import timezone
 from django.contrib.auth.middleware import RemoteUserMiddleware
 from django.core.exceptions import ImproperlyConfigured
@@ -28,10 +29,11 @@ def TimezoneMiddleware(get_response):


 def CacheControlMiddleware(get_response):
+    snapshot_path_re = re.compile(r"^/[^/]+/\\d{8}/[^/]+/[0-9a-fA-F-]{8,36}/")
    def middleware(request):
        response = get_response(request)

-        if '/archive/' in request.path or '/static/' in request.path:
+        if '/archive/' in request.path or '/static/' in request.path or snapshot_path_re.match(request.path):
            policy = 'public' if SERVER_CONFIG.PUBLIC_SNAPSHOTS else 'private'
            response['Cache-Control'] = f'{policy}, max-age=60, stale-while-revalidate=300'
            # print('Set Cache-Control header to', response['Cache-Control'])
--- a/archivebox/core/models.py
+++ b/archivebox/core/models.py
@@ -1296,7 +1296,6 @@ class Snapshot(ModelWithOutputDir, ModelWithConfig, ModelWithNotes, ModelWithHea
                )}

            path = self.archive_path
-            canon = self.canonical_outputs()
            output = ""
            output_template = '<a href="/{}/{}" class="exists-{}" title="{}">{}</a> &nbsp;'

@@ -1313,10 +1312,11 @@ class Snapshot(ModelWithOutputDir, ModelWithConfig, ModelWithNotes, ModelWithHea
                if not icon.strip() and not existing:
                    continue

+                embed_path = result.embed_path() if result else f'{plugin}/'
                output += format_html(
                    output_template,
                    path,
-                    canon.get(plugin, plugin + '/'),
+                    embed_path,
                    str(bool(existing)),
                    plugin,
                    icon
@@ -1402,9 +1402,38 @@ class Snapshot(ModelWithOutputDir, ModelWithConfig, ModelWithNotes, ModelWithHea
            return

    @cached_property
-    def archive_path(self):
+    def legacy_archive_path(self) -> str:
        return f'{CONSTANTS.ARCHIVE_DIR_NAME}/{self.timestamp}'

+    @cached_property
+    def url_path(self) -> str:
+        """URL path matching the current snapshot output_dir layout."""
+        try:
+            rel_path = Path(self.output_dir).resolve().relative_to(CONSTANTS.DATA_DIR)
+        except Exception:
+            return self.legacy_archive_path
+
+        parts = rel_path.parts
+        # New layout: users/<username>/snapshots/<YYYYMMDD>/<domain>/<uuid>/
+        if len(parts) >= 6 and parts[0] == 'users' and parts[2] == 'snapshots':
+            username = parts[1]
+            if username == 'system':
+                username = 'web'
+            date_str = parts[3]
+            domain = parts[4]
+            snapshot_id = parts[5]
+            return f'{username}/{date_str}/{domain}/{snapshot_id}'
+
+        # Legacy layout: archive/<timestamp>/
+        if len(parts) >= 2 and parts[0] == CONSTANTS.ARCHIVE_DIR_NAME:
+            return f'{parts[0]}/{parts[1]}'
+
+        return '/'.join(parts)
+
+    @cached_property
+    def archive_path(self):
+        return self.url_path
+
    @cached_property
    def archive_size(self):
        try:
@@ -1467,8 +1496,8 @@ class Snapshot(ModelWithOutputDir, ModelWithConfig, ModelWithNotes, ModelWithHea
            for pid_file in Path(self.output_dir).glob('**/*.pid'):
                pid_file.unlink(missing_ok=True)

-        # Update all STARTED ArchiveResults from filesystem
-        results = self.archiveresult_set.filter(status=ArchiveResult.StatusChoices.STARTED)
+        # Update all background ArchiveResults from filesystem (in case output arrived late)
+        results = self.archiveresult_set.filter(hook_name__contains='.bg.')
        for ar in results:
            ar.update_from_output()

@@ -1914,153 +1943,6 @@ class Snapshot(ModelWithOutputDir, ModelWithConfig, ModelWithNotes, ModelWithHea
    # Output Path Methods (migrated from Link schema)
    # =========================================================================

-    def canonical_outputs(self) -> Dict[str, Optional[str]]:
-        """
-        Intelligently discover the best output file for each plugin.
-        Uses actual ArchiveResult data and filesystem scanning with smart heuristics.
-        """
-        FAVICON_PROVIDER = 'https://www.google.com/s2/favicons?domain={}'
-
-        # Mimetypes that can be embedded/previewed in an iframe
-        IFRAME_EMBEDDABLE_EXTENSIONS = {
-            'html', 'htm', 'pdf', 'txt', 'md', 'json', 'jsonl',
-            'png', 'jpg', 'jpeg', 'gif', 'webp', 'svg', 'ico',
-            'mp4', 'webm', 'mp3', 'opus', 'ogg', 'wav',
-        }
-
-        MIN_DISPLAY_SIZE = 15_000  # 15KB - filter out tiny files
-        MAX_SCAN_FILES = 50  # Don't scan massive directories
-
-        def find_best_output_in_dir(dir_path: Path, plugin_name: str) -> Optional[str]:
-            """Find the best representative file in a plugin's output directory"""
-            if not dir_path.exists() or not dir_path.is_dir():
-                return None
-
-            candidates = []
-            file_count = 0
-
-            # Special handling for media plugin - look for thumbnails
-            is_media_dir = plugin_name == 'media'
-
-            # Scan for suitable files
-            for file_path in dir_path.rglob('*'):
-                file_count += 1
-                if file_count > MAX_SCAN_FILES:
-                    break
-
-                if file_path.is_dir() or file_path.name.startswith('.'):
-                    continue
-
-                ext = file_path.suffix.lstrip('.').lower()
-                if ext not in IFRAME_EMBEDDABLE_EXTENSIONS:
-                    continue
-
-                try:
-                    size = file_path.stat().st_size
-                except OSError:
-                    continue
-
-                # For media dir, allow smaller image files (thumbnails are often < 15KB)
-                min_size = 5_000 if (is_media_dir and ext in ('png', 'jpg', 'jpeg', 'webp', 'gif')) else MIN_DISPLAY_SIZE
-                if size < min_size:
-                    continue
-
-                # Prefer main files: index.html, output.*, content.*, etc.
-                priority = 0
-                name_lower = file_path.name.lower()
-
-                if is_media_dir:
-                    # Special prioritization for media directories
-                    if any(keyword in name_lower for keyword in ('thumb', 'thumbnail', 'cover', 'poster')):
-                        priority = 200  # Highest priority for thumbnails
-                    elif ext in ('png', 'jpg', 'jpeg', 'webp', 'gif'):
-                        priority = 150  # High priority for any image
-                    elif ext in ('mp4', 'webm', 'mp3', 'opus', 'ogg'):
-                        priority = 100  # Lower priority for actual media files
-                    else:
-                        priority = 50
-                elif 'index' in name_lower:
-                    priority = 100
-                elif name_lower.startswith(('output', 'content', plugin_name)):
-                    priority = 50
-                elif ext in ('html', 'htm', 'pdf'):
-                    priority = 30
-                elif ext in ('png', 'jpg', 'jpeg', 'webp'):
-                    priority = 20
-                else:
-                    priority = 10
-
-                candidates.append((priority, size, file_path))
-
-            if not candidates:
-                return None
-
-            # Sort by priority (desc), then size (desc)
-            candidates.sort(key=lambda x: (x[0], x[1]), reverse=True)
-            best_file = candidates[0][2]
-            return str(best_file.relative_to(Path(self.output_dir)))
-
-        canonical = {
-            'index_path': 'index.html',
-            'google_favicon_path': FAVICON_PROVIDER.format(self.domain),
-            'archivedotorg_path': f'https://web.archive.org/web/{self.base_url}',
-        }
-
-        # Scan each ArchiveResult's output directory for the best file
-        snap_dir = Path(self.output_dir)
-        for result in self.archiveresult_set.filter(status='succeeded'):
-            if not result.output_files and not result.output_str:
-                continue
-
-            # Try to find the best output file for this plugin
-            plugin_dir = snap_dir / result.plugin
-            best_output = None
-
-            # Check output_files first (new field)
-            if result.output_files:
-                first_file = next(iter(result.output_files.keys()), None)
-                if first_file and (plugin_dir / first_file).exists():
-                    best_output = f'{result.plugin}/{first_file}'
-
-            # Fallback to output_str if it looks like a path
-            if not best_output and result.output_str and (snap_dir / result.output_str).exists():
-                best_output = result.output_str
-
-            if not best_output and plugin_dir.exists():
-                # Intelligently find the best file in the plugin's directory
-                best_output = find_best_output_in_dir(plugin_dir, result.plugin)
-
-            if best_output:
-                canonical[f'{result.plugin}_path'] = best_output
-
-        # Also scan top-level for legacy outputs (backwards compatibility)
-        for file_path in snap_dir.glob('*'):
-            if file_path.is_dir() or file_path.name in ('index.html', 'index.json'):
-                continue
-
-            ext = file_path.suffix.lstrip('.').lower()
-            if ext not in IFRAME_EMBEDDABLE_EXTENSIONS:
-                continue
-
-            try:
-                size = file_path.stat().st_size
-                if size >= MIN_DISPLAY_SIZE:
-                    # Add as generic output with stem as key
-                    key = f'{file_path.stem}_path'
-                    if key not in canonical:
-                        canonical[key] = file_path.name
-            except OSError:
-                continue
-
-        if self.is_static:
-            static_path = f'warc/{self.timestamp}'
-            canonical.update({
-                'title': self.basename,
-                'wget_path': static_path,
-            })
-
-        return canonical
-
    def latest_outputs(self, status: Optional[str] = None) -> Dict[str, Any]:
        """Get the latest output that each plugin produced"""
        from archivebox.hooks import get_plugins
@@ -2078,6 +1960,96 @@ class Snapshot(ModelWithOutputDir, ModelWithConfig, ModelWithNotes, ModelWithHea
            latest[plugin] = result.embed_path() if result else None
        return latest

+    def discover_outputs(self) -> list[dict]:
+        """Discover output files from ArchiveResults and filesystem."""
+        from archivebox.misc.util import ts_to_date_str
+
+        ArchiveResult = self.archiveresult_set.model
+        snap_dir = Path(self.output_dir)
+        outputs: list[dict] = []
+        seen: set[str] = set()
+
+        text_exts = ('.json', '.jsonl', '.txt', '.csv', '.tsv', '.xml', '.yml', '.yaml', '.md', '.log')
+
+        def is_metadata_path(path: str | None) -> bool:
+            lower = (path or '').lower()
+            return lower.endswith(text_exts)
+
+        def is_compact_path(path: str | None) -> bool:
+            lower = (path or '').lower()
+            return lower.endswith(text_exts)
+
+        for result in self.archiveresult_set.all().order_by('start_ts'):
+            embed_path = result.embed_path()
+            if not embed_path or embed_path.strip() in ('.', '/', './'):
+                continue
+            abs_path = snap_dir / embed_path
+            if not abs_path.exists():
+                continue
+            if abs_path.is_dir():
+                if not any(p.is_file() for p in abs_path.rglob('*')):
+                    continue
+                size = sum(p.stat().st_size for p in abs_path.rglob('*') if p.is_file())
+            else:
+                size = abs_path.stat().st_size
+            outputs.append({
+                'name': result.plugin,
+                'path': embed_path,
+                'ts': ts_to_date_str(result.end_ts),
+                'size': size or 0,
+                'is_metadata': is_metadata_path(embed_path),
+                'is_compact': is_compact_path(embed_path),
+                'result': result,
+            })
+            seen.add(result.plugin)
+
+        embeddable_exts = {
+            'html', 'htm', 'pdf', 'txt', 'md', 'json', 'jsonl', 'csv', 'tsv',
+            'png', 'jpg', 'jpeg', 'gif', 'webp', 'svg', 'ico',
+            'mp4', 'webm', 'mp3', 'opus', 'ogg', 'wav',
+        }
+
+        for entry in snap_dir.iterdir():
+            if entry.name in ('index.html', 'index.json', 'favicon.ico', 'warc'):
+                continue
+            if entry.is_dir():
+                plugin = entry.name
+                if plugin in seen:
+                    continue
+                best_file = ArchiveResult._find_best_output_file(entry, plugin)
+                if not best_file:
+                    continue
+                rel_path = str(best_file.relative_to(snap_dir))
+                outputs.append({
+                    'name': plugin,
+                    'path': rel_path,
+                    'ts': ts_to_date_str(best_file.stat().st_mtime or 0),
+                    'size': best_file.stat().st_size or 0,
+                    'is_metadata': is_metadata_path(rel_path),
+                    'is_compact': is_compact_path(rel_path),
+                    'result': None,
+                })
+                seen.add(plugin)
+            elif entry.is_file():
+                ext = entry.suffix.lstrip('.').lower()
+                if ext not in embeddable_exts:
+                    continue
+                plugin = entry.stem
+                if plugin in seen:
+                    continue
+                outputs.append({
+                    'name': plugin,
+                    'path': entry.name,
+                    'ts': ts_to_date_str(entry.stat().st_mtime or 0),
+                    'size': entry.stat().st_size or 0,
+                    'is_metadata': is_metadata_path(entry.name),
+                    'is_compact': is_compact_path(entry.name),
+                    'result': None,
+                })
+                seen.add(plugin)
+
+        return outputs
+
    # =========================================================================
    # Serialization Methods
    # =========================================================================
@@ -2114,8 +2086,6 @@ class Snapshot(ModelWithOutputDir, ModelWithConfig, ModelWithNotes, ModelWithHea
            'num_outputs': self.num_outputs,
            'num_failures': self.num_failures,
        }
-        if extended:
-            result['canonical'] = self.canonical_outputs()
        return result

    def to_json_str(self, indent: int = 4) -> str:
@@ -2146,23 +2116,29 @@ class Snapshot(ModelWithOutputDir, ModelWithConfig, ModelWithNotes, ModelWithHea
        SAVE_ARCHIVE_DOT_ORG = config.get('SAVE_ARCHIVE_DOT_ORG', True)
        TITLE_LOADING_MSG = 'Not yet archived...'

-        canonical = self.canonical_outputs()
        preview_priority = [
-            'singlefile_path',
-            'screenshot_path',
-            'wget_path',
-            'dom_path',
-            'pdf_path',
-            'readability_path',
+            'singlefile',
+            'screenshot',
+            'wget',
+            'dom',
+            'pdf',
+            'readability',
        ]
-        best_preview_path = next(
-            (canonical.get(key) for key in preview_priority if canonical.get(key)),
-            canonical.get('index_path', 'index.html'),
-        )
+
+        outputs = self.discover_outputs()
+        outputs_by_plugin = {out['name']: out for out in outputs}
+
+        best_preview_path = 'about:blank'
+        for plugin in preview_priority:
+            out = outputs_by_plugin.get(plugin)
+            if out and out.get('path'):
+                best_preview_path = out['path']
+                break
+
+        if best_preview_path == 'about:blank' and outputs:
+            best_preview_path = outputs[0].get('path') or 'about:blank'
        context = {
            **self.to_dict(extended=True),
-            **{f'{k}_path': v for k, v in canonical.items()},
-            'canonical': {f'{k}_path': v for k, v in canonical.items()},
            'title': htmlencode(self.title or (self.base_url if self.is_archived else TITLE_LOADING_MSG)),
            'url_str': htmlencode(urldecode(self.base_url)),
            'archive_url': urlencode(f'warc/{self.timestamp}' or (self.domain if self.is_archived else '')) or 'about:blank',
@@ -2175,6 +2151,7 @@ class Snapshot(ModelWithOutputDir, ModelWithConfig, ModelWithNotes, ModelWithHea
            'SAVE_ARCHIVE_DOT_ORG': SAVE_ARCHIVE_DOT_ORG,
            'PREVIEW_ORIGINALS': SERVER_CONFIG.PREVIEW_ORIGINALS,
            'best_preview_path': best_preview_path,
+            'archiveresults': outputs,
        }
        rendered_html = render_to_string('snapshot.html', context)
        atomic_write(str(Path(out_dir) / CONSTANTS.HTML_INDEX_FILENAME), rendered_html)
@@ -2496,6 +2473,61 @@ class ArchiveResult(ModelWithOutputDir, ModelWithConfig, ModelWithNotes, ModelWi
    def output_exists(self) -> bool:
        return os.path.exists(Path(self.snapshot_dir) / self.plugin)

+    @staticmethod
+    def _find_best_output_file(dir_path: Path, plugin_name: str | None = None) -> Optional[Path]:
+        if not dir_path.exists() or not dir_path.is_dir():
+            return None
+
+        embeddable_exts = {
+            'html', 'htm', 'pdf', 'txt', 'md', 'json', 'jsonl', 'csv', 'tsv',
+            'png', 'jpg', 'jpeg', 'gif', 'webp', 'svg', 'ico',
+            'mp4', 'webm', 'mp3', 'opus', 'ogg', 'wav',
+        }
+
+        for name in ('index.html', 'index.htm'):
+            candidate = dir_path / name
+            if candidate.exists() and candidate.is_file():
+                return candidate
+
+        candidates = []
+        file_count = 0
+        max_scan = 200
+        plugin_lower = (plugin_name or '').lower()
+        for file_path in dir_path.rglob('*'):
+            file_count += 1
+            if file_count > max_scan:
+                break
+            if file_path.is_dir() or file_path.name.startswith('.'):
+                continue
+            ext = file_path.suffix.lstrip('.').lower()
+            if ext not in embeddable_exts:
+                continue
+            try:
+                size = file_path.stat().st_size
+            except OSError:
+                continue
+            name_lower = file_path.name.lower()
+            priority = 0
+            if name_lower.startswith('index'):
+                priority = 100
+            elif plugin_lower and name_lower.startswith(('output', 'content', plugin_lower)):
+                priority = 60
+            elif ext in ('html', 'htm', 'pdf'):
+                priority = 40
+            elif ext in ('png', 'jpg', 'jpeg', 'webp', 'svg', 'gif', 'ico'):
+                priority = 30
+            elif ext in ('json', 'jsonl', 'txt', 'md', 'csv', 'tsv'):
+                priority = 20
+            else:
+                priority = 10
+            candidates.append((priority, size, file_path))
+
+        if not candidates:
+            return None
+
+        candidates.sort(key=lambda x: (x[0], x[1]), reverse=True)
+        return candidates[0][2]
+
    def embed_path(self) -> Optional[str]:
        """
        Get the relative path to the embeddable output file for this result.
@@ -2503,25 +2535,45 @@ class ArchiveResult(ModelWithOutputDir, ModelWithConfig, ModelWithNotes, ModelWi
        Returns the first file from output_files if set, otherwise tries to
        find a reasonable default based on the plugin type.
        """
-        # Check output_files dict for primary output
+        snapshot_dir = Path(self.snapshot_dir)
+        plugin_dir = snapshot_dir / self.plugin
+
+        # Fallback: treat output_str as a file path only if it exists on disk
+        if self.output_str:
+            try:
+                output_path = Path(self.output_str)
+
+                if output_path.is_absolute():
+                    # If absolute and within snapshot dir, normalize to relative
+                    if snapshot_dir in output_path.parents and output_path.exists():
+                        return str(output_path.relative_to(snapshot_dir))
+                else:
+                    # If relative, prefer plugin-prefixed path, then direct path
+                    if (plugin_dir / output_path).exists():
+                        return f'{self.plugin}/{output_path}'
+                    if output_path.name in ('index.html', 'index.json') and output_path.parent == Path('.'):
+                        return None
+                    if (snapshot_dir / output_path).exists():
+                        return str(output_path)
+            except Exception:
+                pass
+
+        # Check output_files dict for primary output (ignore non-output files)
        if self.output_files:
-            # Return first file from output_files (dict preserves insertion order)
-            first_file = next(iter(self.output_files.keys()), None)
-            if first_file:
+            ignored = {'stdout.log', 'stderr.log', 'hook.pid', 'listener.pid', 'cmd.sh'}
+            output_candidates = [
+                f for f in self.output_files.keys()
+                if Path(f).name not in ignored
+            ]
+            first_file = output_candidates[0] if output_candidates else None
+            if first_file and (plugin_dir / first_file).exists():
                return f'{self.plugin}/{first_file}'

-        # Fallback: check output_str if it looks like a file path
-        if self.output_str and ('/' in self.output_str or '.' in self.output_str):
-            return self.output_str
+        best_file = self._find_best_output_file(plugin_dir, self.plugin)
+        if best_file:
+            return str(best_file.relative_to(snapshot_dir))

-        # Try to find output file based on plugin's canonical output path
-        canonical = self.snapshot.canonical_outputs()
-        plugin_key = f'{self.plugin}_path'
-        if plugin_key in canonical:
-            return canonical[plugin_key]
-
-        # Fallback to plugin directory
-        return f'{self.plugin}/'
+        return None

    def create_output_dir(self):
        output_dir = Path(self.snapshot_dir) / self.plugin
@@ -2779,7 +2831,7 @@ class ArchiveResult(ModelWithOutputDir, ModelWithConfig, ModelWithNotes, ModelWi
                self.output_str = 'Hook did not output ArchiveResult record'

        # Walk filesystem and populate output_files, output_size, output_mimetypes
-        exclude_names = {'stdout.log', 'stderr.log', 'hook.pid', 'listener.pid'}
+        exclude_names = {'stdout.log', 'stderr.log', 'hook.pid', 'listener.pid', 'cmd.sh'}
        mime_sizes = defaultdict(int)
        total_size = 0
        output_files = {}
--- a/archivebox/core/settings_logging.py
+++ b/archivebox/core/settings_logging.py
@@ -48,6 +48,19 @@ class CustomOutboundWebhookLogFormatter(logging.Formatter):
        result = super().format(record)
        return result.replace('HTTP Request: ', 'OutboundWebhook: ')

+class StripANSIColorCodesFilter(logging.Filter):
+    _ansi_re = re.compile(r'\x1b\[[0-9;]*m')
+    _bare_re = re.compile(r'\[[0-9;]*m')
+
+    def filter(self, record) -> bool:
+        msg = record.getMessage()
+        if isinstance(msg, str) and ('\x1b[' in msg or '[m' in msg):
+            msg = self._ansi_re.sub('', msg)
+            msg = self._bare_re.sub('', msg)
+            record.msg = msg
+            record.args = ()
+        return True
+

 ERROR_LOG = tempfile.NamedTemporaryFile().name

@@ -87,6 +100,9 @@ SETTINGS_LOGGING = {
        "noisyrequestsfilter": {
            "()": NoisyRequestsFilter,
        },
+        "stripansi": {
+            "()": StripANSIColorCodesFilter,
+        },
        "require_debug_false": {
            "()": "django.utils.log.RequireDebugFalse",
        },
@@ -101,7 +117,7 @@ SETTINGS_LOGGING = {
            "level": "DEBUG",
            "markup": False,
            "rich_tracebacks": False,  # Use standard Python tracebacks (no frame/box)
-            "filters": ["noisyrequestsfilter"],
+            "filters": ["noisyrequestsfilter", "stripansi"],
        },
        "logfile": {
            "level": "INFO",
@@ -110,7 +126,7 @@ SETTINGS_LOGGING = {
            "maxBytes": 1024 * 1024 * 25,  # 25 MB
            "backupCount": 10,
            "formatter": "rich",
-            "filters": ["noisyrequestsfilter"],
+            "filters": ["noisyrequestsfilter", "stripansi"],
        },
        "outbound_webhooks": {
            "class": "rich.logging.RichHandler",
--- a/archivebox/core/templatetags/core_tags.py
+++ b/archivebox/core/templatetags/core_tags.py
@@ -1,8 +1,10 @@
 from django import template
 from django.contrib.admin.templatetags.base import InclusionAdminNode
 from django.utils.safestring import mark_safe
+from django.utils.html import escape

 from typing import Union
+from pathlib import Path

 from archivebox.hooks import (
    get_plugin_icon, get_plugin_template, get_plugin_name,
@@ -57,15 +59,18 @@ def plugin_icon(plugin: str) -> str:

    Usage: {% plugin_icon "screenshot" %}
    """
-    return mark_safe(get_plugin_icon(plugin))
+    icon_html = get_plugin_icon(plugin)
+    return mark_safe(
+        f'<span class="abx-plugin-icon" style="display:inline-flex; width:20px; height:20px; align-items:center; justify-content:center;">{icon_html}</span>'
+    )


@register.simple_tag(takes_context=True)
-def plugin_thumbnail(context, result) -> str:
+def plugin_card(context, result) -> str:
    """
-    Render the thumbnail template for an archive result.
+    Render the card template for an archive result.

-    Usage: {% plugin_thumbnail result %}
+    Usage: {% plugin_card result %}

    Context variables passed to template:
        - result: ArchiveResult object
@@ -74,46 +79,97 @@ def plugin_thumbnail(context, result) -> str:
        - plugin: Plugin base name
    """
    plugin = get_plugin_name(result.plugin)
-    template_str = get_plugin_template(plugin, 'thumbnail')
+    template_str = get_plugin_template(plugin, 'card')

-    if not template_str:
-        return ''
+    # Use embed_path() for the display path
+    output_path = result.embed_path() if hasattr(result, 'embed_path') else ''

-    # Use embed_path() for the display path (includes canonical paths)
-    output_path = result.embed_path() if hasattr(result, 'embed_path') else (result.output_str or '')
+    icon_html = get_plugin_icon(plugin)
+
+    output_lower = (output_path or '').lower()
+    text_preview_exts = ('.json', '.jsonl', '.txt', '.csv', '.tsv', '.xml', '.yml', '.yaml', '.md', '.log')
+    force_text_preview = output_lower.endswith(text_preview_exts)

    # Create a mini template and render it with context
    try:
-        tpl = template.Template(template_str)
-        ctx = template.Context({
-            'result': result,
-            'snapshot': result.snapshot,
-            'output_path': output_path,
-            'plugin': plugin,
-        })
-        rendered = tpl.render(ctx)
-        # Only return non-empty content (strip whitespace to check)
-        if rendered.strip():
-            return mark_safe(rendered)
-        return ''
+        if template_str and output_path and str(output_path).strip() not in ('.', '/', './') and not force_text_preview:
+            tpl = template.Template(template_str)
+            ctx = template.Context({
+                'result': result,
+                'snapshot': result.snapshot,
+                'output_path': output_path,
+                'plugin': plugin,
+                'plugin_icon': icon_html,
+            })
+            rendered = tpl.render(ctx)
+            # Only return non-empty content (strip whitespace to check)
+            if rendered.strip():
+                return mark_safe(rendered)
    except Exception:
-        return ''
+        pass
+
+    if force_text_preview and output_path and str(output_path).strip() not in ('.', '/', './'):
+        output_file = Path(output_path)
+        if not output_file.is_absolute():
+            output_file = Path(result.snapshot_dir) / output_path
+        try:
+            output_file = output_file.resolve()
+            snap_dir = Path(result.snapshot_dir).resolve()
+            if snap_dir not in output_file.parents and output_file != snap_dir:
+                output_file = None
+        except Exception:
+            output_file = None
+        if output_file and output_file.exists() and output_file.is_file():
+            try:
+                with output_file.open('rb') as f:
+                    raw = f.read(4096)
+                text = raw.decode('utf-8', errors='replace').strip()
+                if text:
+                    lines = text.splitlines()[:6]
+                    snippet = '\n'.join(lines)
+                    escaped = escape(snippet)
+                    preview = (
+                        f'<div class="thumbnail-text" data-plugin="{plugin}" data-compact="1">'
+                        f'<div class="thumbnail-text-header">'
+                        f'<span class="thumbnail-compact-icon">{icon_html}</span>'
+                        f'<span class="thumbnail-text-title">{plugin}</span>'
+                        f'</div>'
+                        f'<pre class="thumbnail-text-pre">{escaped}</pre>'
+                        f'</div>'
+                    )
+                    return mark_safe(preview)
+            except Exception:
+                pass
+
+    if output_lower.endswith(text_preview_exts):
+        fallback_label = 'text'
+    else:
+        fallback_label = 'output'
+
+    fallback = (
+        f'<div class="thumbnail-compact" data-plugin="{plugin}" data-compact="1">'
+        f'<span class="thumbnail-compact-icon">{icon_html}</span>'
+        f'<span class="thumbnail-compact-label">{plugin}</span>'
+        f'<span class="thumbnail-compact-meta">{fallback_label}</span>'
+        f'</div>'
+    )
+    return mark_safe(fallback)


@register.simple_tag(takes_context=True)
-def plugin_embed(context, result) -> str:
+def plugin_full(context, result) -> str:
    """
-    Render the embed iframe template for an archive result.
+    Render the full template for an archive result.

-    Usage: {% plugin_embed result %}
+    Usage: {% plugin_full result %}
    """
    plugin = get_plugin_name(result.plugin)
-    template_str = get_plugin_template(plugin, 'embed')
+    template_str = get_plugin_template(plugin, 'full')

    if not template_str:
        return ''

-    output_path = result.embed_path() if hasattr(result, 'embed_path') else (result.output_str or '')
+    output_path = result.embed_path() if hasattr(result, 'embed_path') else ''

    try:
        tpl = template.Template(template_str)
@@ -132,36 +188,6 @@ def plugin_embed(context, result) -> str:
        return ''


-@register.simple_tag(takes_context=True)
-def plugin_fullscreen(context, result) -> str:
-    """
-    Render the fullscreen template for an archive result.
-
-    Usage: {% plugin_fullscreen result %}
-    """
-    plugin = get_plugin_name(result.plugin)
-    template_str = get_plugin_template(plugin, 'fullscreen')
-
-    if not template_str:
-        return ''
-
-    output_path = result.embed_path() if hasattr(result, 'embed_path') else (result.output_str or '')
-
-    try:
-        tpl = template.Template(template_str)
-        ctx = template.Context({
-            'result': result,
-            'snapshot': result.snapshot,
-            'output_path': output_path,
-            'plugin': plugin,
-        })
-        rendered = tpl.render(ctx)
-        # Only return non-empty content (strip whitespace to check)
-        if rendered.strip():
-            return mark_safe(rendered)
-        return ''
-    except Exception:
-        return ''


@register.filter
--- a/archivebox/core/urls.py
+++ b/archivebox/core/urls.py
@@ -8,7 +8,7 @@ from django.views.generic.base import RedirectView
 from archivebox.misc.serve_static import serve_static

 from archivebox.core.admin_site import archivebox_admin
-from archivebox.core.views import HomepageView, SnapshotView, PublicIndexView, AddView, HealthCheckView, live_progress_view
+from archivebox.core.views import HomepageView, SnapshotView, SnapshotPathView, PublicIndexView, AddView, HealthCheckView, live_progress_view

 from archivebox.workers.views import JobsDashboardView

@@ -32,6 +32,8 @@ urlpatterns = [
    
    path('archive/', RedirectView.as_view(url='/')),
    path('archive/<path:path>', SnapshotView.as_view(), name='Snapshot'),
+    re_path(r'^(?P<username>[^/]+)/(?P<date>\d{4}(?:\d{2})?(?:\d{2})?)/(?P<url>https?://.*)$', SnapshotPathView.as_view(), name='snapshot-path-url'),
+    re_path(r'^(?P<username>[^/]+)/(?P<date>\d{4}(?:\d{2})?(?:\d{2})?)/(?P<domain>[^/]+)(?:/(?P<snapshot_id>[0-9a-fA-F-]{8,36})(?:/(?P<path>.*))?)?$', SnapshotPathView.as_view(), name='snapshot-path'),

    path('admin/core/snapshot/add/', RedirectView.as_view(url='/add/')),
    path('add/', AddView.as_view(), name='add'),
--- a/archivebox/core/views.py
+++ b/archivebox/core/views.py
@@ -1,7 +1,6 @@
 __package__ = 'archivebox.core'

 import os
-import sys
 from django.utils import timezone
 import inspect
 from typing import Callable, get_type_hints
@@ -26,7 +25,7 @@ import archivebox
 from archivebox.config import CONSTANTS, CONSTANTS_CONFIG, DATA_DIR, VERSION
 from archivebox.config.common import SHELL_CONFIG, SERVER_CONFIG, ARCHIVING_CONFIG
 from archivebox.config.configset import get_flat_config, get_config, get_all_configs
-from archivebox.misc.util import base_url, htmlencode, ts_to_date_str
+from archivebox.misc.util import base_url, htmlencode, ts_to_date_str, urldecode
 from archivebox.misc.serve_static import serve_static_with_byterange_support
 from archivebox.misc.logging_util import printable_filesize
 from archivebox.search import query_search_index
@@ -52,70 +51,44 @@ class HomepageView(View):
 class SnapshotView(View):
    # render static html index from filesystem archive/<timestamp>/index.html

+    @staticmethod
+    def find_snapshots_for_url(path: str):
+        """Return a queryset of snapshots matching a URL-ish path."""
+        normalized = path
+        if path.startswith(('http://', 'https://')):
+            # try exact match on full url / ID first
+            qs = Snapshot.objects.filter(Q(url=path) | Q(id__icontains=path))
+            if qs.exists():
+                return qs
+            normalized = path.split('://', 1)[1]
+
+        # try exact match on full url / ID (without scheme)
+        qs = Snapshot.objects.filter(
+            Q(url='http://' + normalized) | Q(url='https://' + normalized) | Q(id__icontains=normalized)
+        )
+        if qs.exists():
+            return qs
+
+        # fall back to match on exact base_url
+        base = base_url(normalized)
+        qs = Snapshot.objects.filter(
+            Q(url='http://' + base) | Q(url='https://' + base)
+        )
+        if qs.exists():
+            return qs
+
+        # fall back to matching base_url as prefix
+        return Snapshot.objects.filter(
+            Q(url__startswith='http://' + base) | Q(url__startswith='https://' + base)
+        )
+
    @staticmethod
    def render_live_index(request, snapshot):
        TITLE_LOADING_MSG = 'Not yet archived...'

-        # Dict of plugin -> ArchiveResult object
-        archiveresult_objects = {}
-        # Dict of plugin -> result info dict (for template compatibility)
-        archiveresults = {}
-
-        results = snapshot.archiveresult_set.all()
-
-        for result in results:
-            embed_path = result.embed_path()
-            abs_path = result.snapshot_dir / (embed_path or 'None')
-
-            if (result.status == 'succeeded'
-                and embed_path
-                and os.access(abs_path, os.R_OK)
-                and abs_path.exists()):
-                if os.path.isdir(abs_path) and not any(abs_path.glob('*.*')):
-                    continue
-
-                # Store the full ArchiveResult object for template tags
-                archiveresult_objects[result.plugin] = result
-
-                result_info = {
-                    'name': result.plugin,
-                    'path': embed_path,
-                    'ts': ts_to_date_str(result.end_ts),
-                    'size': abs_path.stat().st_size or '?',
-                    'result': result,  # Include the full object for template tags
-                }
-                archiveresults[result.plugin] = result_info
-
-        # Use canonical_outputs for intelligent discovery
-        # This method now scans ArchiveResults and uses smart heuristics
-        canonical = snapshot.canonical_outputs()
-
-        # Add any newly discovered outputs from canonical_outputs to archiveresults
+        outputs = snapshot.discover_outputs()
+        archiveresults = {out['name']: out for out in outputs}
        snap_dir = Path(snapshot.output_dir)
-        for key, path in canonical.items():
-            if not key.endswith('_path') or not path or path.startswith('http'):
-                continue
-
-            plugin_name = key.replace('_path', '')
-            if plugin_name in archiveresults:
-                continue  # Already have this from ArchiveResult
-
-            file_path = snap_dir / path
-            if not file_path.exists() or not file_path.is_file():
-                continue
-
-            try:
-                file_size = file_path.stat().st_size
-                if file_size >= 15_000:  # Only show files > 15KB
-                    archiveresults[plugin_name] = {
-                        'name': plugin_name,
-                        'path': path,
-                        'ts': ts_to_date_str(file_path.stat().st_mtime or 0),
-                        'size': file_size,
-                        'result': None,
-                    }
-            except OSError:
-                continue

        # Get available extractor plugins from hooks (sorted by numeric prefix for ordering)
        # Convert to base names for display ordering
@@ -131,7 +104,7 @@ class SnapshotView(View):
        preferred_types = tuple(preview_priority + [p for p in all_plugins if p not in preview_priority])
        all_types = preferred_types + tuple(result_type for result_type in archiveresults.keys() if result_type not in preferred_types)

-        best_result = {'path': 'None', 'result': None}
+        best_result = {'path': 'about:blank', 'result': None}
        for result_type in preferred_types:
            if result_type in archiveresults:
                best_result = archiveresults[result_type]
@@ -146,7 +119,6 @@ class SnapshotView(View):

        context = {
            **snapshot_info,
-            **snapshot_info.get('canonical', {}),
            'title': htmlencode(
                snapshot.title
                or (snapshot.base_url if snapshot.is_archived else TITLE_LOADING_MSG)
@@ -188,6 +160,14 @@ class SnapshotView(View):
            try:
                try:
                    snapshot = Snapshot.objects.get(Q(timestamp=slug) | Q(id__startswith=slug))
+                    canonical_base = snapshot.url_path
+                    if canonical_base != snapshot.legacy_archive_path:
+                        target_path = f'/{canonical_base}/{archivefile or "index.html"}'
+                        query = request.META.get('QUERY_STRING')
+                        if query:
+                            target_path = f'{target_path}?{query}'
+                        return redirect(target_path)
+
                    if archivefile == 'index.html':
                        # if they requested snapshot index, serve live rendered template instead of static html
                        response = self.render_live_index(request, snapshot)
@@ -221,9 +201,9 @@ class SnapshotView(View):
            except Snapshot.MultipleObjectsReturned:
                snapshot_hrefs = mark_safe('<br/>').join(
                    format_html(
-                        '{} <a href="/archive/{}/index.html"><b><code>{}</code></b></a> {} <b>{}</b>',
+                        '{} <a href="/{}/index.html"><b><code>{}</code></b></a> {} <b>{}</b>',
                        snap.bookmarked_at.strftime('%Y-%m-%d %H:%M:%S'),
-                        snap.timestamp,
+                        snap.archive_path,
                        snap.timestamp,
                        snap.url,
                        snap.title_stripped[:64] or '',
@@ -259,9 +239,9 @@ class SnapshotView(View):
                            #'</script>'
                            '</head><body>'
                            '<center><br/><br/><br/>'
-                            f'Snapshot <a href="/archive/{snapshot.timestamp}/index.html" target="_top"><b><code>[{snapshot.timestamp}]</code></b></a>: <a href="{snapshot.url}" target="_blank" rel="noreferrer">{snapshot.url}</a><br/>'
+                            f'Snapshot <a href="/{snapshot.archive_path}/index.html" target="_top"><b><code>[{snapshot.timestamp}]</code></b></a>: <a href="{snapshot.url}" target="_blank" rel="noreferrer">{snapshot.url}</a><br/>'
                            f'was queued on {str(snapshot.bookmarked_at).split(".")[0]}, '
-                            f'but no files have been saved yet in:<br/><b><a href="/archive/{snapshot.timestamp}/" target="_top"><code>{snapshot.timestamp}</code></a><code>/'
+                            f'but no files have been saved yet in:<br/><b><a href="/{snapshot.archive_path}/" target="_top"><code>{snapshot.timestamp}</code></a><code>/'
                            '{}'
                            f'</code></b><br/><br/>'
                            'It\'s possible {} '
@@ -270,8 +250,8 @@ class SnapshotView(View):
                            f'<code style="user-select: all; color: #333">archivebox update -t timestamp {snapshot.timestamp}</code></pre><br/><br/>'
                            '<div class="text-align: left; width: 100%; max-width: 400px">'
                            '<i><b>Next steps:</i></b><br/>'
-                            f'- list all the <a href="/archive/{snapshot.timestamp}/" target="_top">Snapshot files <code>.*</code></a><br/>'
-                            f'- view the <a href="/archive/{snapshot.timestamp}/index.html" target="_top">Snapshot <code>./index.html</code></a><br/>'
+                            f'- list all the <a href="/{snapshot.archive_path}/" target="_top">Snapshot files <code>.*</code></a><br/>'
+                            f'- view the <a href="/{snapshot.archive_path}/index.html" target="_top">Snapshot <code>./index.html</code></a><br/>'
                            f'- go to the <a href="/admin/core/snapshot/{snapshot.pk}/change/" target="_top">Snapshot admin</a> to edit<br/>'
                            f'- go to the <a href="/admin/core/snapshot/?id__exact={snapshot.id}" target="_top">Snapshot actions</a> to re-archive<br/>'
                            '- or return to <a href="/" target="_top">the main index...</a></div>'
@@ -288,22 +268,9 @@ class SnapshotView(View):
        # slug is a URL
        try:
            try:
-                # try exact match on full url / ID first
-                snapshot = Snapshot.objects.get(
-                    Q(url='http://' + path) | Q(url='https://' + path) | Q(id__icontains=path)
-                )
+                snapshot = SnapshotView.find_snapshots_for_url(path).get()
            except Snapshot.DoesNotExist:
-                # fall back to match on exact base_url
-                try:
-                    snapshot = Snapshot.objects.get(
-                        Q(url='http://' + base_url(path)) | Q(url='https://' + base_url(path))
-                    )
-                except Snapshot.DoesNotExist:
-                    # fall back to matching base_url as prefix
-                    snapshot = Snapshot.objects.get(
-                        Q(url__startswith='http://' + base_url(path)) | Q(url__startswith='https://' + base_url(path))
-                    )
-            return redirect(f'/archive/{snapshot.timestamp}/index.html')
+                raise
        except Snapshot.DoesNotExist:
            return HttpResponse(
                format_html(
@@ -322,20 +289,18 @@ class SnapshotView(View):
                status=404,
            )
        except Snapshot.MultipleObjectsReturned:
+            snapshots = SnapshotView.find_snapshots_for_url(path)
            snapshot_hrefs = mark_safe('<br/>').join(
                format_html(
-                    '{} <code style="font-size: 0.8em">{}</code> <a href="/archive/{}/index.html"><b><code>{}</code></b></a> {} <b>{}</b>',
+                    '{} <code style="font-size: 0.8em">{}</code> <a href="/{}/index.html"><b><code>{}</code></b></a> {} <b>{}</b>',
                    snap.bookmarked_at.strftime('%Y-%m-%d %H:%M:%S'),
                    str(snap.id)[:8],
-                    snap.timestamp,
+                    snap.archive_path,
                    snap.timestamp,
                    snap.url,
                    snap.title_stripped[:64] or '',
                )
-                for snap in Snapshot.objects.filter(
-                    Q(url__startswith='http://' + base_url(path)) | Q(url__startswith='https://' + base_url(path))
-                    | Q(id__icontains=path)
-                ).only('url', 'timestamp', 'title', 'bookmarked_at').order_by('-bookmarked_at')
+                for snap in snapshots.only('url', 'timestamp', 'title', 'bookmarked_at').order_by('-bookmarked_at')
            )
            return HttpResponse(
                format_html(
@@ -353,6 +318,108 @@ class SnapshotView(View):
                status=404,
            )

+        target_path = f'/{snapshot.archive_path}/index.html'
+        query = request.META.get('QUERY_STRING')
+        if query:
+            target_path = f'{target_path}?{query}'
+        return redirect(target_path)
+
+
+class SnapshotPathView(View):
+    """Serve snapshots by the new URL scheme: /<username>/<YYYYMMDD>/<domain>/<uuid>/..."""
+
+    def get(self, request, username: str, date: str, domain: str | None = None, snapshot_id: str | None = None, path: str = "", url: str | None = None):
+        if not request.user.is_authenticated and not SERVER_CONFIG.PUBLIC_SNAPSHOTS:
+            return redirect(f'/admin/login/?next={request.path}')
+
+        if username == 'system':
+            return redirect(request.path.replace('/system/', '/web/', 1))
+
+        requested_url = url
+        if not requested_url and domain and domain.startswith(('http://', 'https://')):
+            requested_url = domain
+
+        snapshot = None
+        if snapshot_id:
+            try:
+                snapshot = Snapshot.objects.get(pk=snapshot_id)
+            except Snapshot.DoesNotExist:
+                try:
+                    snapshot = Snapshot.objects.get(id__startswith=snapshot_id)
+                except Snapshot.DoesNotExist:
+                    snapshot = None
+                except Snapshot.MultipleObjectsReturned:
+                    snapshot = Snapshot.objects.filter(id__startswith=snapshot_id).first()
+        else:
+            # fuzzy lookup by date + domain/url (most recent)
+            username_lookup = 'system' if username == 'web' else username
+            if requested_url:
+                qs = SnapshotView.find_snapshots_for_url(requested_url).filter(crawl__created_by__username=username_lookup)
+            else:
+                qs = Snapshot.objects.filter(crawl__created_by__username=username_lookup)
+
+            try:
+                if len(date) == 4:
+                    qs = qs.filter(created_at__year=int(date))
+                elif len(date) == 6:
+                    qs = qs.filter(created_at__year=int(date[:4]), created_at__month=int(date[4:6]))
+                elif len(date) == 8:
+                    qs = qs.filter(
+                        created_at__year=int(date[:4]),
+                        created_at__month=int(date[4:6]),
+                        created_at__day=int(date[6:8]),
+                    )
+            except ValueError:
+                pass
+
+            if requested_url:
+                snapshot = qs.order_by('-created_at', '-bookmarked_at', '-timestamp').first()
+            else:
+                requested_domain = domain or ''
+                if requested_domain.startswith(('http://', 'https://')):
+                    requested_domain = Snapshot.extract_domain_from_url(requested_domain)
+                else:
+                    requested_domain = Snapshot.extract_domain_from_url(f'https://{requested_domain}')
+
+                # Prefer exact domain matches
+                matches = [s for s in qs.order_by('-created_at', '-bookmarked_at') if Snapshot.extract_domain_from_url(s.url) == requested_domain]
+                snapshot = matches[0] if matches else qs.order_by('-created_at', '-bookmarked_at', '-timestamp').first()
+
+        if not snapshot:
+            return HttpResponse(
+                format_html(
+                    (
+                        '<center><br/><br/><br/>'
+                        'No Snapshots match the given id or url: <code>{}</code><br/><br/><br/>'
+                        'Return to the <a href="/" target="_top">Main Index</a>'
+                        '</center>'
+                    ),
+                    snapshot_id or requested_url or domain,
+                ),
+                content_type="text/html",
+                status=404,
+            )
+
+        canonical_base = snapshot.url_path
+        requested_base = f'{username}/{date}/{domain or url or ""}'
+        if snapshot_id:
+            requested_base = f'{requested_base}/{snapshot_id}'
+        if canonical_base != requested_base:
+            target = f'/{canonical_base}/{path or "index.html"}'
+            query = request.META.get('QUERY_STRING')
+            if query:
+                target = f'{target}?{query}'
+            return redirect(target)
+
+        archivefile = path or "index.html"
+
+        if archivefile == "index.html":
+            return SnapshotView.render_live_index(request, snapshot)
+
+        return serve_static_with_byterange_support(
+            request, archivefile, document_root=snapshot.output_dir, show_indexes=True,
+        )
+

 class PublicIndexView(ListView):
    template_name = 'public_index.html'
@@ -592,7 +659,7 @@ def live_progress_view(request):
                        'snapshot_id': str(ar.snapshot_id),
                        'snapshot_url': ar.snapshot.url[:60] if ar.snapshot else '',
                        'embed_path': embed,
-                        'archive_path': f'/archive/{ar.snapshot.timestamp}/{embed}' if ar.snapshot else '',
+                        'archive_path': f'/{ar.snapshot.archive_path}/{embed}' if ar.snapshot else '',
                        'end_ts': ar.end_ts.isoformat() if ar.end_ts else None,
                    })

--- a/archivebox/crawls/admin.py
+++ b/archivebox/crawls/admin.py
@@ -71,8 +71,8 @@ def render_snapshots_list(snapshots_qs, limit=20):
                                 color: {color}; background: {bg};">{status}</span>
                </td>
                <td style="padding: 6px 8px; white-space: nowrap;">
-                    <a href="/archive/{snapshot.timestamp}/" style="text-decoration: none;">
-                        <img src="/archive/{snapshot.timestamp}/favicon.ico"
+                    <a href="/{snapshot.archive_path}/" style="text-decoration: none;">
+                        <img src="/{snapshot.archive_path}/favicon.ico"
                             style="width: 16px; height: 16px; vertical-align: middle; margin-right: 4px;"
                             onerror="this.style.display='none'"/>
                    </a>
--- a/archivebox/hooks.py
+++ b/archivebox/hooks.py
@@ -940,9 +940,8 @@ def get_plugin_special_config(plugin_name: str, config: Dict[str, Any]) -> Dict[
 #     archivebox/plugins/<plugin_name>/
 #         templates/
 #             icon.html          # Icon for admin table view (small inline HTML)
-#             thumbnail.html     # Preview thumbnail for snapshot cards
-#             embed.html         # Iframe embed content for main preview
-#             fullscreen.html    # Fullscreen view template
+#             card.html          # Preview card for snapshot header
+#             full.html          # Fullscreen view template
 #
 # Template context variables available:
 #     {{ result }}         - ArchiveResult object
@@ -953,21 +952,22 @@ def get_plugin_special_config(plugin_name: str, config: Dict[str, Any]) -> Dict[

 # Default templates used when plugin doesn't provide one
 DEFAULT_TEMPLATES = {
-    'icon': '''<span title="{{ plugin }}">{{ icon }}</span>''',
-    'thumbnail': '''
-        <img src="{{ output_path }}"
-             alt="{{ plugin }} output"
-             style="max-width: 100%; max-height: 100px; object-fit: cover;"
-             onerror="this.style.display='none'">
+    'icon': '''
+        <span title="{{ plugin }}" style="display:inline-flex; width:20px; height:20px; align-items:center; justify-content:center;">
+            {{ icon }}
+        </span>
    ''',
-    'embed': '''
+    'card': '''
        <iframe src="{{ output_path }}"
+                class="card-img-top"
                style="width: 100%; height: 100%; border: none;"
-                sandbox="allow-same-origin allow-scripts">
+                sandbox="allow-same-origin allow-scripts allow-forms"
+                loading="lazy">
        </iframe>
    ''',
-    'fullscreen': '''
+    'full': '''
        <iframe src="{{ output_path }}"
+                class="full-page-iframe"
                style="width: 100%; height: 100vh; border: none;"
                sandbox="allow-same-origin allow-scripts allow-forms">
        </iframe>
@@ -981,7 +981,7 @@ def get_plugin_template(plugin: str, template_name: str, fallback: bool = True)

    Args:
        plugin: Plugin name (e.g., 'screenshot', '15_singlefile')
-        template_name: One of 'icon', 'thumbnail', 'embed', 'fullscreen'
+        template_name: One of 'icon', 'card', 'full'
        fallback: If True, return default template if plugin template not found

    Returns:
@@ -1050,7 +1050,7 @@ def discover_plugin_templates() -> Dict[str, Dict[str, str]]:

    Returns:
        Dict mapping plugin names to dicts of template_name -> template_path.
-        e.g., {'screenshot': {'icon': '/path/to/icon.html', 'thumbnail': '/path/to/thumbnail.html'}}
+        e.g., {'screenshot': {'icon': '/path/to/icon.html', 'card': '/path/to/card.html'}}
    """
    templates: Dict[str, Dict[str, str]] = {}

@@ -1068,7 +1068,7 @@ def discover_plugin_templates() -> Dict[str, Dict[str, str]]:

            plugin_templates = {}
            for template_file in templates_dir.glob('*.html'):
-                template_name = template_file.stem  # icon, thumbnail, embed, fullscreen
+                template_name = template_file.stem  # icon, card, full
                plugin_templates[template_name] = str(template_file)

            if plugin_templates:
--- a/archivebox/machine/models.py
+++ b/archivebox/machine/models.py
@@ -237,8 +237,8 @@ class Binary(ModelWithHealthStats, ModelWithStateMachine):
    sha256 = models.CharField(max_length=64, default='', null=False, blank=True)

    # State machine fields
-    status = models.CharField(max_length=16, choices=StatusChoices.choices, default=StatusChoices.QUEUED, db_index=True)
-    retry_at = models.DateTimeField(default=timezone.now, null=True, blank=True, db_index=True,
+    status = ModelWithStateMachine.StatusField(choices=StatusChoices.choices, default=StatusChoices.QUEUED, max_length=16)
+    retry_at = ModelWithStateMachine.RetryAtField(default=timezone.now,
        help_text="When to retry this binary installation")

    # Health stats
@@ -246,6 +246,7 @@ class Binary(ModelWithHealthStats, ModelWithStateMachine):
    num_uses_succeeded = models.PositiveIntegerField(default=0)

    state_machine_name: str = 'archivebox.machine.models.BinaryMachine'
+    active_state: str = StatusChoices.QUEUED

    objects: BinaryManager = BinaryManager()

--- a/archivebox/plugins/accessibility/templates/thumbnail.html
+++ b/archivebox/plugins/accessibility/templates/thumbnail.html
--- a/archivebox/plugins/archivedotorg/templates/thumbnail.html
+++ b/archivebox/plugins/archivedotorg/templates/thumbnail.html
--- a/archivebox/plugins/chrome/on_Crawl__90_chrome_launch.bg.js
+++ b/archivebox/plugins/chrome/on_Crawl__90_chrome_launch.bg.js
@@ -49,6 +49,111 @@ const OUTPUT_DIR = '.';
 let chromePid = null;
 let browserInstance = null;

+function parseCookiesTxt(contents) {
+    const cookies = [];
+    let skipped = 0;
+
+    for (const rawLine of contents.split(/\r?\n/)) {
+        const line = rawLine.trim();
+        if (!line) continue;
+
+        let httpOnly = false;
+        let dataLine = line;
+
+        if (dataLine.startsWith('#HttpOnly_')) {
+            httpOnly = true;
+            dataLine = dataLine.slice('#HttpOnly_'.length);
+        } else if (dataLine.startsWith('#')) {
+            continue;
+        }
+
+        const parts = dataLine.split('\t');
+        if (parts.length < 7) {
+            skipped += 1;
+            continue;
+        }
+
+        const [domainRaw, includeSubdomainsRaw, pathRaw, secureRaw, expiryRaw, name, value] = parts;
+        if (!name || !domainRaw) {
+            skipped += 1;
+            continue;
+        }
+
+        const includeSubdomains = (includeSubdomainsRaw || '').toUpperCase() === 'TRUE';
+        let domain = domainRaw;
+        if (includeSubdomains && !domain.startsWith('.')) domain = `.${domain}`;
+        if (!includeSubdomains && domain.startsWith('.')) domain = domain.slice(1);
+
+        const cookie = {
+            name,
+            value,
+            domain,
+            path: pathRaw || '/',
+            secure: (secureRaw || '').toUpperCase() === 'TRUE',
+            httpOnly,
+        };
+
+        const expires = parseInt(expiryRaw, 10);
+        if (!isNaN(expires) && expires > 0) {
+            cookie.expires = expires;
+        }
+
+        cookies.push(cookie);
+    }
+
+    return { cookies, skipped };
+}
+
+async function importCookiesFromFile(browser, cookiesFile, userDataDir) {
+    if (!cookiesFile) return;
+
+    if (!fs.existsSync(cookiesFile)) {
+        console.error(`[!] Cookies file not found: ${cookiesFile}`);
+        return;
+    }
+
+    let contents = '';
+    try {
+        contents = fs.readFileSync(cookiesFile, 'utf-8');
+    } catch (e) {
+        console.error(`[!] Failed to read COOKIES_TXT_FILE: ${e.message}`);
+        return;
+    }
+
+    const { cookies, skipped } = parseCookiesTxt(contents);
+    if (cookies.length === 0) {
+        console.error('[!] No cookies found to import');
+        return;
+    }
+
+    console.error(`[*] Importing ${cookies.length} cookies from ${cookiesFile}...`);
+    if (skipped) {
+        console.error(`[*] Skipped ${skipped} malformed cookie line(s)`);
+    }
+    if (!userDataDir) {
+        console.error('[!] CHROME_USER_DATA_DIR not set; cookies will not persist beyond this session');
+    }
+
+    const page = await browser.newPage();
+    const client = await page.target().createCDPSession();
+    await client.send('Network.enable');
+
+    const chunkSize = 200;
+    let imported = 0;
+    for (let i = 0; i < cookies.length; i += chunkSize) {
+        const chunk = cookies.slice(i, i + chunkSize);
+        try {
+            await client.send('Network.setCookies', { cookies: chunk });
+            imported += chunk.length;
+        } catch (e) {
+            console.error(`[!] Failed to import cookies ${i + 1}-${i + chunk.length}: ${e.message}`);
+        }
+    }
+
+    await page.close();
+    console.error(`[+] Imported ${imported}/${cookies.length} cookies`);
+}
+
 // Parse command line arguments
 function parseArgs() {
    const args = {};
@@ -118,10 +223,14 @@ async function main() {
        // Load installed extensions
        const extensionsDir = getExtensionsDir();
        const userDataDir = getEnv('CHROME_USER_DATA_DIR');
+        const cookiesFile = getEnv('COOKIES_TXT_FILE') || getEnv('COOKIES_FILE');

        if (userDataDir) {
            console.error(`[*] Using user data dir: ${userDataDir}`);
        }
+        if (cookiesFile) {
+            console.error(`[*] Using cookies file: ${cookiesFile}`);
+        }

        const installedExtensions = [];
        const extensionPaths = [];
@@ -179,6 +288,9 @@ async function main() {
        });
        browserInstance = browser;

+        // Import cookies into Chrome profile at crawl start
+        await importCookiesFromFile(browser, cookiesFile, userDataDir);
+
        // Get actual extension IDs from chrome://extensions page
        if (extensionPaths.length > 0) {
            await new Promise(r => setTimeout(r, 2000));
--- a/archivebox/plugins/chrome/tests/test_chrome.py
+++ b/archivebox/plugins/chrome/tests/test_chrome.py
@@ -38,6 +38,82 @@ from archivebox.plugins.chrome.tests.chrome_test_helpers import (
    CHROME_NAVIGATE_HOOK,
 )

+def _get_cookies_via_cdp(port: int, env: dict) -> list[dict]:
+    node_script = r"""
+const http = require('http');
+const WebSocket = require('ws');
+const port = process.env.CDP_PORT;
+
+function getTargets() {
+  return new Promise((resolve, reject) => {
+    const req = http.get(`http://127.0.0.1:${port}/json/list`, (res) => {
+      let data = '';
+      res.on('data', (chunk) => (data += chunk));
+      res.on('end', () => {
+        try {
+          resolve(JSON.parse(data));
+        } catch (e) {
+          reject(e);
+        }
+      });
+    });
+    req.on('error', reject);
+  });
+}
+
+(async () => {
+  const targets = await getTargets();
+  const pageTarget = targets.find(t => t.type === 'page') || targets[0];
+  if (!pageTarget) {
+    console.error('No page target found');
+    process.exit(2);
+  }
+
+  const ws = new WebSocket(pageTarget.webSocketDebuggerUrl);
+  const timer = setTimeout(() => {
+    console.error('Timeout waiting for cookies');
+    process.exit(3);
+  }, 10000);
+
+  ws.on('open', () => {
+    ws.send(JSON.stringify({ id: 1, method: 'Network.getAllCookies' }));
+  });
+
+  ws.on('message', (data) => {
+    const msg = JSON.parse(data);
+    if (msg.id === 1) {
+      clearTimeout(timer);
+      ws.close();
+      if (!msg.result || !msg.result.cookies) {
+        console.error('No cookies in response');
+        process.exit(4);
+      }
+      process.stdout.write(JSON.stringify(msg.result.cookies));
+      process.exit(0);
+    }
+  });
+
+  ws.on('error', (err) => {
+    console.error(String(err));
+    process.exit(5);
+  });
+})().catch((err) => {
+  console.error(String(err));
+  process.exit(1);
+});
+"""
+
+    result = subprocess.run(
+        ['node', '-e', node_script],
+        capture_output=True,
+        text=True,
+        timeout=30,
+        env=env | {'CDP_PORT': str(port)},
+    )
+    assert result.returncode == 0, f"Failed to read cookies via CDP: {result.stderr}\nStdout: {result.stdout}"
+    return json.loads(result.stdout or '[]')
+
+
@pytest.fixture(scope="session", autouse=True)
 def ensure_chromium_and_puppeteer_installed(tmp_path_factory):
    """Ensure Chromium and puppeteer are installed before running tests."""
@@ -197,6 +273,77 @@ def test_chrome_launch_and_tab_creation():
            pass


+def test_cookies_imported_on_launch():
+    """Integration test: COOKIES_TXT_FILE is imported at crawl start."""
+    with tempfile.TemporaryDirectory() as tmpdir:
+        crawl_dir = Path(tmpdir) / 'crawl'
+        crawl_dir.mkdir()
+        chrome_dir = crawl_dir / 'chrome'
+        chrome_dir.mkdir()
+
+        cookies_file = Path(tmpdir) / 'cookies.txt'
+        cookies_file.write_text(
+            '\n'.join([
+                '# Netscape HTTP Cookie File',
+                '# https://curl.se/docs/http-cookies.html',
+                '# This file was generated by a test',
+                '',
+                'example.com\tTRUE\t/\tFALSE\t2147483647\tabx_test_cookie\thello',
+                '',
+            ])
+        )
+
+        profile_dir = Path(tmpdir) / 'profile'
+        env = get_test_env()
+        env.update({
+            'CHROME_HEADLESS': 'true',
+            'CHROME_USER_DATA_DIR': str(profile_dir),
+            'COOKIES_TXT_FILE': str(cookies_file),
+        })
+
+        chrome_launch_process = subprocess.Popen(
+            ['node', str(CHROME_LAUNCH_HOOK), '--crawl-id=test-crawl-cookies'],
+            cwd=str(chrome_dir),
+            stdout=subprocess.PIPE,
+            stderr=subprocess.PIPE,
+            text=True,
+            env=env
+        )
+
+        for _ in range(15):
+            if (chrome_dir / 'port.txt').exists():
+                break
+            time.sleep(1)
+
+        assert (chrome_dir / 'port.txt').exists(), "port.txt should exist"
+        chrome_pid = int((chrome_dir / 'chrome.pid').read_text().strip())
+        port = int((chrome_dir / 'port.txt').read_text().strip())
+
+        cookie_found = False
+        for _ in range(15):
+            cookies = _get_cookies_via_cdp(port, env)
+            cookie_found = any(
+                c.get('name') == 'abx_test_cookie' and c.get('value') == 'hello'
+                for c in cookies
+            )
+            if cookie_found:
+                break
+            time.sleep(1)
+
+        assert cookie_found, "Imported cookie should be present in Chrome session"
+
+        # Cleanup
+        try:
+            chrome_launch_process.send_signal(signal.SIGTERM)
+            chrome_launch_process.wait(timeout=5)
+        except:
+            pass
+        try:
+            os.kill(chrome_pid, signal.SIGKILL)
+        except OSError:
+            pass
+
+
 def test_chrome_navigation():
    """Integration test: Navigate to a URL."""
    with tempfile.TemporaryDirectory() as tmpdir:
--- a/archivebox/plugins/consolelog/templates/thumbnail.html
+++ b/archivebox/plugins/consolelog/templates/thumbnail.html
--- a/archivebox/plugins/dns/templates/thumbnail.html
+++ b/archivebox/plugins/dns/templates/thumbnail.html
--- a/archivebox/plugins/dom/templates/thumbnail.html
+++ b/archivebox/plugins/dom/templates/thumbnail.html
--- a/archivebox/plugins/favicon/templates/thumbnail.html
+++ b/archivebox/plugins/favicon/templates/thumbnail.html
--- a/archivebox/plugins/forumdl/templates/thumbnail.html
+++ b/archivebox/plugins/forumdl/templates/thumbnail.html
--- a/archivebox/plugins/forumdl/templates/embed.html
+++ b/archivebox/plugins/forumdl/templates/embed.html
@@ -1,40 +0,0 @@
-<!-- Embedded forum view - renders JSONL forum posts -->
-<div class="extractor-embed forumdl-embed" style="width: 100%; max-width: 900px; margin: 0 auto; background: #1a1a1a; padding: 20px; border-radius: 8px;">
-    <div style="text-align: center; padding: 15px 0; border-bottom: 1px solid #333; margin-bottom: 20px;">
-        <span style="font-size: 32px;">💬</span>
-        <h3 style="margin: 10px 0; color: #fff; font-size: 18px;">Forum Thread</h3>
-    </div>
-    <div id="forum-posts" style="max-height: 500px; overflow-y: auto; color: #ddd;"></div>
-    <script>
-        (async function() {
-            try {
-                const response = await fetch('{{ output_path }}');
-                const text = await response.text();
-                const posts = text.trim().split('\n').map(line => JSON.parse(line));
-                const container = document.getElementById('forum-posts');
-
-                posts.forEach(post => {
-                    const postDiv = document.createElement('div');
-                    postDiv.style.cssText = 'background: #2a2a2a; padding: 15px; margin-bottom: 15px; border-radius: 5px; border-left: 3px solid #4a9eff;';
-
-                    const author = post.author || 'Anonymous';
-                    const date = post.date ? new Date(post.date).toLocaleString() : '';
-                    const title = post.title || '';
-                    const content = post.content || post.body || '';
-
-                    postDiv.innerHTML = `
-                        <div style="display: flex; justify-content: space-between; margin-bottom: 10px; padding-bottom: 8px; border-bottom: 1px solid #444;">
-                            <strong style="color: #4a9eff;">${author}</strong>
-                            <span style="color: #888; font-size: 12px;">${date}</span>
-                        </div>
-                        ${title ? `<h4 style="margin: 0 0 10px 0; color: #fff;">${title}</h4>` : ''}
-                        <div style="color: #ccc; line-height: 1.5;">${content}</div>
-                    `;
-                    container.appendChild(postDiv);
-                });
-            } catch(e) {
-                document.getElementById('forum-posts').innerHTML = '<p style="color: #888;">Error loading forum posts</p>';
-            }
-        })();
-    </script>
-</div>
--- a/archivebox/plugins/forumdl/templates/fullscreen.html
+++ b/archivebox/plugins/forumdl/templates/fullscreen.html
--- a/archivebox/plugins/gallerydl/templates/thumbnail.html
+++ b/archivebox/plugins/gallerydl/templates/thumbnail.html
--- a/archivebox/plugins/gallerydl/templates/embed.html
+++ b/archivebox/plugins/gallerydl/templates/embed.html
@@ -1,11 +0,0 @@
-<!-- Embedded gallery view - shows first image with link to full gallery -->
-<div class="extractor-embed gallerydl-embed" style="width: 100%; max-width: 800px; margin: 0 auto; background: #1a1a1a; padding: 20px;">
-    <img src="{{ output_path }}"
-         style="width: 100%; max-height: 600px; object-fit: contain;"
-         alt="Gallery image"
-         onerror="this.style.display='none'; this.nextElementSibling.style.display='flex';">
-    <div style="display: none; flex-direction: column; align-items: center; color: #888; padding: 40px;">
-        <span style="font-size: 64px;">🖼️</span>
-        <span style="margin-top: 10px;">Gallery downloaded</span>
-    </div>
-</div>
--- a/archivebox/plugins/gallerydl/templates/fullscreen.html
+++ b/archivebox/plugins/gallerydl/templates/fullscreen.html
--- a/archivebox/plugins/git/templates/thumbnail.html
+++ b/archivebox/plugins/git/templates/thumbnail.html
--- a/archivebox/plugins/mercury/templates/thumbnail.html
+++ b/archivebox/plugins/mercury/templates/thumbnail.html
--- a/archivebox/plugins/papersdl/templates/thumbnail.html
+++ b/archivebox/plugins/papersdl/templates/thumbnail.html
--- a/archivebox/plugins/papersdl/templates/embed.html
+++ b/archivebox/plugins/papersdl/templates/embed.html
@@ -1,15 +0,0 @@
-<!-- Embedded paper view - shows PDF viewer -->
-<div class="extractor-embed papersdl-embed" style="width: 100%; max-width: 900px; margin: 0 auto; background: #1a1a1a; padding: 20px; border-radius: 8px;">
-    <div style="text-align: center; padding: 15px 0; border-bottom: 1px solid #333; margin-bottom: 20px;">
-        <span style="font-size: 32px;">📄</span>
-        <h3 style="margin: 10px 0; color: #fff; font-size: 18px;">Scientific Paper</h3>
-    </div>
-    <div style="width: 100%; height: 500px; background: #2a2a2a; border-radius: 5px; overflow: hidden;">
-        <embed src="{{ output_path }}" type="application/pdf" width="100%" height="100%" />
-    </div>
-    <div style="margin-top: 15px; text-align: center;">
-        <a href="{{ output_path }}" download style="color: #4a9eff; text-decoration: none; padding: 10px 20px; background: #2a2a2a; border-radius: 5px; display: inline-block;">
-            Download PDF
-        </a>
-    </div>
-</div>
--- a/archivebox/plugins/papersdl/templates/fullscreen.html
+++ b/archivebox/plugins/papersdl/templates/fullscreen.html
--- a/archivebox/plugins/parse_html_urls/on_Snapshot__70_parse_html_urls.py
+++ b/archivebox/plugins/parse_html_urls/on_Snapshot__70_parse_html_urls.py
@@ -271,12 +271,11 @@ def main(url: str, snapshot_id: str = None, crawl_id: str = None, depth: int = 0
        records.append(record)
        print(json.dumps(record))

-    if records:
-        URLS_FILE.write_text('\n'.join(json.dumps(r) for r in records) + '\n')
+    URLS_FILE.write_text('\n'.join(json.dumps(r) for r in records) + ('\n' if records else ''))

    # Emit ArchiveResult record to mark completion
    status = 'succeeded' if urls_found else 'skipped'
-    output_str = f'Found {len(urls_found)} URLs' if urls_found else 'No URLs found'
+    output_str = URLS_FILE.name
    ar_record = {
        'type': 'ArchiveResult',
        'status': status,
--- a/archivebox/plugins/parse_html_urls/tests/test_parse_html_urls.py
+++ b/archivebox/plugins/parse_html_urls/tests/test_parse_html_urls.py
@@ -57,7 +57,7 @@ class TestParseHtmlUrls:
        )

        assert result.returncode == 0
-        assert 'Found 3 URLs' in result.stderr
+        assert 'urls.jsonl' in result.stderr

        # Parse Snapshot records from stdout
        lines = [line for line in result.stdout.strip().split('\n') if line.strip() and '"type": "Snapshot"' in line]
@@ -78,6 +78,11 @@ class TestParseHtmlUrls:
        assert '"type": "ArchiveResult"' in result.stdout
        assert '"status": "succeeded"' in result.stdout

+        urls_file = tmp_path / 'urls.jsonl'
+        assert urls_file.exists(), "urls.jsonl not created"
+        file_lines = [line for line in urls_file.read_text().splitlines() if line.strip()]
+        assert len(file_lines) == 3, f"Expected 3 urls.jsonl entries, got {len(file_lines)}"
+
    def test_ignores_non_http_schemes(self, tmp_path):
        """Test that non-http schemes are ignored."""
        input_file = tmp_path / 'page.html'
@@ -194,7 +199,7 @@ class TestParseHtmlUrls:
        )

        assert result.returncode == 0
-        assert 'No URLs found' in result.stderr
+        assert 'urls.jsonl' in result.stderr
        assert '"status": "skipped"' in result.stdout

    def test_handles_malformed_html(self, tmp_path):
--- a/archivebox/plugins/parse_jsonl_urls/on_Snapshot__74_parse_jsonl_urls.py
+++ b/archivebox/plugins/parse_jsonl_urls/on_Snapshot__74_parse_jsonl_urls.py
@@ -18,6 +18,7 @@ Supports various field names for URL, title, timestamp, and tags.
 import json
 import os
 import sys
+from pathlib import Path
 from datetime import datetime
 from html import unescape
 from urllib.parse import urlparse
@@ -25,6 +26,7 @@ from urllib.parse import urlparse
 import rich_click as click

 PLUGIN_NAME = 'parse_jsonl_urls'
+URLS_FILE = Path('urls.jsonl')


 def parse_bookmarked_at(link: dict) -> str | None:
@@ -188,9 +190,12 @@ def main(url: str, snapshot_id: str = None, crawl_id: str = None, depth: int = 0
    for entry in urls_found:
        print(json.dumps(entry))

+    # Write urls.jsonl to disk for crawl system
+    URLS_FILE.write_text('\n'.join(json.dumps(r) for r in urls_found) + ('\n' if urls_found else ''))
+
    # Emit ArchiveResult record to mark completion
    status = 'succeeded' if urls_found else 'skipped'
-    output_str = f'Found {len(urls_found)} URLs, {len(all_tags)} tags' if urls_found else 'No URLs found'
+    output_str = URLS_FILE.name
    ar_record = {
        'type': 'ArchiveResult',
        'status': status,
--- a/archivebox/plugins/parse_jsonl_urls/tests/test_parse_jsonl_urls.py
+++ b/archivebox/plugins/parse_jsonl_urls/tests/test_parse_jsonl_urls.py
@@ -32,7 +32,7 @@ class TestParseJsonlUrls:
        )

        assert result.returncode == 0
-        assert 'Found 3 URLs' in result.stdout
+        assert 'urls.jsonl' in result.stderr or 'urls.jsonl' in result.stdout

        # Output goes to stdout (JSONL)
        lines = [line for line in result.stdout.strip().split('\n') if line.strip() and '\"type\": \"Snapshot\"' in line]
@@ -195,7 +195,7 @@ class TestParseJsonlUrls:
        )

        assert result.returncode == 0
-        assert 'No URLs found' in result.stderr
+        assert 'urls.jsonl' in result.stderr
        assert '"status": "skipped"' in result.stdout

    def test_exits_1_when_file_not_found(self, tmp_path):
--- a/archivebox/plugins/parse_netscape_urls/on_Snapshot__73_parse_netscape_urls.py
+++ b/archivebox/plugins/parse_netscape_urls/on_Snapshot__73_parse_netscape_urls.py
@@ -16,6 +16,7 @@ import json
 import os
 import re
 import sys
+from pathlib import Path
 from datetime import datetime, timezone
 from html import unescape
 from urllib.parse import urlparse
@@ -23,6 +24,7 @@ from urllib.parse import urlparse
 import rich_click as click

 PLUGIN_NAME = 'parse_netscape_urls'
+URLS_FILE = Path('urls.jsonl')

 # Constants for timestamp epoch detection
 UNIX_EPOCH = 0  # 1970-01-01 00:00:00 UTC
@@ -232,9 +234,12 @@ def main(url: str, snapshot_id: str = None, crawl_id: str = None, depth: int = 0
    for entry in urls_found:
        print(json.dumps(entry))

+    # Write urls.jsonl to disk for crawl system
+    URLS_FILE.write_text('\n'.join(json.dumps(r) for r in urls_found) + ('\n' if urls_found else ''))
+
    # Emit ArchiveResult record to mark completion
    status = 'succeeded' if urls_found else 'skipped'
-    output_str = f'Found {len(urls_found)} URLs, {len(all_tags)} tags' if urls_found else 'No bookmarks found'
+    output_str = URLS_FILE.name
    ar_record = {
        'type': 'ArchiveResult',
        'status': status,
--- a/archivebox/plugins/parse_netscape_urls/tests/test_parse_netscape_urls.py
+++ b/archivebox/plugins/parse_netscape_urls/tests/test_parse_netscape_urls.py
@@ -37,7 +37,7 @@ class TestParseNetscapeUrls:
        )

        assert result.returncode == 0
-        assert 'Found 3 URLs' in result.stdout
+        assert 'urls.jsonl' in result.stderr or 'urls.jsonl' in result.stdout

        # Output goes to stdout (JSONL)
        lines = [line for line in result.stdout.strip().split('\n') if line.strip() and '\"type\": \"Snapshot\"' in line]
@@ -135,7 +135,7 @@ class TestParseNetscapeUrls:
        )

        assert result.returncode == 0
-        assert 'No bookmarks found' in result.stderr
+        assert 'urls.jsonl' in result.stderr
        assert '"status": "skipped"' in result.stdout

    def test_exits_1_when_file_not_found(self, tmp_path):
--- a/archivebox/plugins/parse_netscape_urls/tests/test_parse_netscape_urls_comprehensive.py
+++ b/archivebox/plugins/parse_netscape_urls/tests/test_parse_netscape_urls_comprehensive.py
@@ -935,7 +935,7 @@ class TestEdgeCases:
        )

        assert result.returncode == 0
-        assert 'Found 1000 URLs' in result.stdout
+        assert 'urls.jsonl' in result.stderr or 'urls.jsonl' in result.stdout

        # Output goes to stdout (JSONL) - get all JSONL records
        all_lines = [line for line in result.stdout.strip().split('\n') if line.strip() and line.startswith('{')]
--- a/archivebox/plugins/parse_rss_urls/on_Snapshot__72_parse_rss_urls.py
+++ b/archivebox/plugins/parse_rss_urls/on_Snapshot__72_parse_rss_urls.py
@@ -16,6 +16,7 @@ Examples:
 import json
 import os
 import sys
+from pathlib import Path
 from datetime import datetime, timezone
 from html import unescape
 from time import mktime
@@ -24,6 +25,7 @@ from urllib.parse import urlparse
 import rich_click as click

 PLUGIN_NAME = 'parse_rss_urls'
+URLS_FILE = Path('urls.jsonl')

 try:
    import feedparser
@@ -140,9 +142,12 @@ def main(url: str, snapshot_id: str = None, crawl_id: str = None, depth: int = 0
    for entry in urls_found:
        print(json.dumps(entry))

+    # Write urls.jsonl to disk for crawl system
+    URLS_FILE.write_text('\n'.join(json.dumps(r) for r in urls_found) + ('\n' if urls_found else ''))
+
    # Emit ArchiveResult record to mark completion
    status = 'succeeded' if urls_found else 'skipped'
-    output_str = f'Found {len(urls_found)} URLs, {len(all_tags)} tags' if urls_found else 'No URLs found'
+    output_str = URLS_FILE.name
    ar_record = {
        'type': 'ArchiveResult',
        'status': status,
--- a/archivebox/plugins/parse_rss_urls/tests/test_parse_rss_urls.py
+++ b/archivebox/plugins/parse_rss_urls/tests/test_parse_rss_urls.py
@@ -66,7 +66,7 @@ class TestParseRssUrls:
        )

        assert result.returncode == 0
-        assert 'Found 2 URLs' in result.stdout
+        assert 'urls.jsonl' in result.stderr or 'urls.jsonl' in result.stdout

        # Output goes to stdout (JSONL)
        lines = [line for line in result.stdout.strip().split('\n') if line.strip() and '\"type\": \"Snapshot\"' in line]
@@ -134,7 +134,7 @@ class TestParseRssUrls:
        )

        assert result.returncode == 0
-        assert 'No URLs found' in result.stderr
+        assert 'urls.jsonl' in result.stderr
        assert '"status": "skipped"' in result.stdout

    def test_exits_1_when_file_not_found(self, tmp_path):
--- a/archivebox/plugins/parse_rss_urls/tests/test_parse_rss_urls_comprehensive.py
+++ b/archivebox/plugins/parse_rss_urls/tests/test_parse_rss_urls_comprehensive.py
@@ -882,7 +882,7 @@ class TestEdgeCases:
        )

        assert result.returncode == 0
-        assert 'Found 100 URLs' in result.stdout
+        assert 'urls.jsonl' in result.stderr or 'urls.jsonl' in result.stdout

        # Output goes to stdout (JSONL)
        lines = [line for line in result.stdout.strip().split('\n') if line.strip()]
--- a/archivebox/plugins/parse_txt_urls/on_Snapshot__71_parse_txt_urls.py
+++ b/archivebox/plugins/parse_txt_urls/on_Snapshot__71_parse_txt_urls.py
@@ -26,6 +26,7 @@ from urllib.request import urlopen
 import rich_click as click

 PLUGIN_NAME = 'parse_txt_urls'
+URLS_FILE = Path('urls.jsonl')

 # URL regex from archivebox/misc/util.py
 # https://mathiasbynens.be/demo/url-regex
@@ -127,6 +128,7 @@ def main(url: str, snapshot_id: str = None, crawl_id: str = None, depth: int = 0
            urls_found.add(cleaned_url)

    # Emit Snapshot records to stdout (JSONL)
+    records = []
    for found_url in sorted(urls_found):
        record = {
            'type': 'Snapshot',
@@ -138,11 +140,13 @@ def main(url: str, snapshot_id: str = None, crawl_id: str = None, depth: int = 0
            record['parent_snapshot_id'] = snapshot_id
        if crawl_id:
            record['crawl_id'] = crawl_id
+        records.append(record)
        print(json.dumps(record))

    # Emit ArchiveResult record to mark completion
+    URLS_FILE.write_text('\n'.join(json.dumps(r) for r in records) + ('\n' if records else ''))
    status = 'succeeded' if urls_found else 'skipped'
-    output_str = f'Found {len(urls_found)} URLs' if urls_found else 'No URLs found'
+    output_str = URLS_FILE.name
    ar_record = {
        'type': 'ArchiveResult',
        'status': status,
--- a/archivebox/plugins/parse_txt_urls/tests/test_parse_txt_urls.py
+++ b/archivebox/plugins/parse_txt_urls/tests/test_parse_txt_urls.py
@@ -32,7 +32,7 @@ https://www.iana.org/domains/reserved
        )

        assert result.returncode == 0, f"Failed: {result.stderr}"
-        assert 'Found 3 URLs' in result.stderr
+        assert 'urls.jsonl' in result.stderr

        # Parse Snapshot records from stdout
        lines = [line for line in result.stdout.strip().split('\n') if line.strip() and '"type": "Snapshot"' in line]
@@ -113,7 +113,7 @@ Also see https://github.com/user/repo for the code.
        )

        assert result.returncode == 0
-        assert 'No URLs found' in result.stderr
+        assert 'urls.jsonl' in result.stderr
        assert '"status": "skipped"' in result.stdout

    def test_exits_1_when_file_not_found(self, tmp_path):
--- a/archivebox/plugins/pdf/templates/thumbnail.html
+++ b/archivebox/plugins/pdf/templates/thumbnail.html
--- a/archivebox/plugins/pdf/templates/embed.html
+++ b/archivebox/plugins/pdf/templates/embed.html
@@ -1,5 +0,0 @@
-<!-- PDF embed - full PDF viewer -->
-<embed src="{{ output_path }}#toolbar=1&navpanes=1"
-       type="application/pdf"
-       class="extractor-embed pdf-embed"
-       style="width: 100%; height: 100%; min-height: 500px;">
--- a/archivebox/plugins/pdf/templates/fullscreen.html
+++ b/archivebox/plugins/pdf/templates/fullscreen.html
--- a/archivebox/plugins/readability/on_Snapshot__56_readability.py
+++ b/archivebox/plugins/readability/on_Snapshot__56_readability.py
@@ -31,6 +31,7 @@ PLUGIN_NAME = 'readability'
 BIN_NAME = 'readability-extractor'
 BIN_PROVIDERS = 'npm,env'
 OUTPUT_DIR = '.'
+OUTPUT_FILE = 'content.html'


 def get_env(name: str, default: str = '') -> str:
@@ -130,11 +131,11 @@ def extract_readability(url: str, binary: str) -> tuple[bool, str | None, str]:
        if not text_content and not html_content:
            return False, None, 'No content extracted'

-        (output_dir / 'content.html').write_text(html_content, encoding='utf-8')
+        (output_dir / OUTPUT_FILE).write_text(html_content, encoding='utf-8')
        (output_dir / 'content.txt').write_text(text_content, encoding='utf-8')
        (output_dir / 'article.json').write_text(json.dumps(result_json, indent=2), encoding='utf-8')

-        return True, OUTPUT_DIR, ''
+        return True, OUTPUT_FILE, ''

    except subprocess.TimeoutExpired:
        return False, None, f'Timed out after {timeout} seconds'
--- a/archivebox/plugins/readability/templates/thumbnail.html
+++ b/archivebox/plugins/readability/templates/thumbnail.html
--- a/archivebox/plugins/readability/templates/full.html
+++ b/archivebox/plugins/readability/templates/full.html
@@ -0,0 +1,6 @@
+<!-- Readability fullscreen - show extracted article HTML -->
+<iframe class="full-page-iframe"
+        src="{{ output_path }}"
+        name="preview"
+        sandbox="allow-same-origin allow-top-navigation-by-user-activation allow-scripts allow-forms">
+</iframe>
--- a/archivebox/plugins/redirects/templates/thumbnail.html
+++ b/archivebox/plugins/redirects/templates/thumbnail.html
--- a/archivebox/plugins/responses/templates/thumbnail.html
+++ b/archivebox/plugins/responses/templates/thumbnail.html
--- a/archivebox/plugins/screenshot/templates/thumbnail.html
+++ b/archivebox/plugins/screenshot/templates/thumbnail.html
--- a/archivebox/plugins/screenshot/templates/embed.html
+++ b/archivebox/plugins/screenshot/templates/embed.html
@@ -1,5 +0,0 @@
-<!-- Screenshot embed - full image view -->
-<img src="{{ output_path }}"
-     alt="Screenshot of page"
-     class="extractor-embed screenshot-embed"
-     style="max-width: 100%; height: auto;">
--- a/archivebox/plugins/screenshot/templates/fullscreen.html
+++ b/archivebox/plugins/screenshot/templates/fullscreen.html
--- a/archivebox/plugins/seo/templates/thumbnail.html
+++ b/archivebox/plugins/seo/templates/thumbnail.html
--- a/archivebox/plugins/singlefile/templates/thumbnail.html
+++ b/archivebox/plugins/singlefile/templates/thumbnail.html
--- a/archivebox/plugins/ssl/templates/thumbnail.html
+++ b/archivebox/plugins/ssl/templates/thumbnail.html
--- a/archivebox/plugins/staticfile/templates/thumbnail.html
+++ b/archivebox/plugins/staticfile/templates/thumbnail.html
--- a/archivebox/plugins/title/on_Snapshot__54_title.js
+++ b/archivebox/plugins/title/on_Snapshot__54_title.js
@@ -263,7 +263,7 @@ async function main() {
    const archiveResult = {
        type: 'ArchiveResult',
        status,
-        output_str: extractedTitle || error || '',
+        output_str: output || error || '',
    };
    console.log(JSON.stringify(archiveResult));

--- a/archivebox/plugins/wget/templates/thumbnail.html
+++ b/archivebox/plugins/wget/templates/thumbnail.html
--- a/archivebox/plugins/ytdlp/templates/thumbnail.html
+++ b/archivebox/plugins/ytdlp/templates/thumbnail.html
--- a/archivebox/plugins/ytdlp/templates/embed.html
+++ b/archivebox/plugins/ytdlp/templates/embed.html
@@ -1,9 +0,0 @@
-<!-- YT-DLP embed - video/audio player -->
-<div class="extractor-embed ytdlp-embed" style="width: 100%; height: 100%; min-height: 400px; background: #1a1a1a; display: flex; align-items: center; justify-content: center;">
-    <video src="{{ output_path }}"
-           style="max-width: 100%; max-height: 100%;"
-           controls
-           preload="metadata">
-        Your browser does not support the video tag.
-    </video>
-</div>
--- a/archivebox/plugins/ytdlp/templates/fullscreen.html
+++ b/archivebox/plugins/ytdlp/templates/fullscreen.html
--- a/archivebox/templates/core/index_row.html
+++ b/archivebox/templates/core/index_row.html
@@ -6,12 +6,12 @@
    </td>
    <td class="title-col" style="opacity: {% if link.title %}1{% else %}0.3{% endif %}" title="{{link.title|default:'Not yet archived...'}}">
        {% if link.is_archived %}
-            <a href="/archive/{{link.timestamp}}/index.html"><img src="/archive/{{link.timestamp}}/favicon.ico" onerror="this.style.display='none'" class="link-favicon" decoding="async"></a>
+            <a href="/{{link.archive_path}}/index.html"><img src="/{{link.archive_path}}/favicon.ico" onerror="this.style.display='none'" class="link-favicon" decoding="async"></a>
        {% else %}
-            <a href="/archive/{{link.timestamp}}/index.html"><img src="{% static 'spinner.gif' %}" onerror="this.style.display='none'" class="link-favicon" decoding="async" style="height: 15px"></a>
+            <a href="/{{link.archive_path}}/index.html"><img src="{% static 'spinner.gif' %}" onerror="this.style.display='none'" class="link-favicon" decoding="async" style="height: 15px"></a>
        {% endif %}

-        <a href="/archive/{{link.timestamp}}/index.html" title="{{link.title|default:'Not yet archived...'}}">
+        <a href="/{{link.archive_path}}/index.html" title="{{link.title|default:'Not yet archived...'}}">
            <span data-title-for="{{link.url}}" data-archived="{{link.is_archived}}">
                {{link.title|default:'Loading...'|truncatechars:128}}
            </span>
@@ -29,7 +29,7 @@
            {% if link.icons  %}
                {{link.icons}}&nbsp; <small style="float:right; opacity: 0.5">{{link.num_outputs}}</small>
            {% else %}
-                <a href="/archive/{{link.timestamp}}/index.html">
+                <a href="/{{link.archive_path}}/index.html">
                    📄 &nbsp;
                    {{link.num_outputs}} <img src="{% static 'spinner.gif' %}" onerror="this.style.display='none'" class="files-spinner" decoding="async" style="height: 15px"/>
                </a>
--- a/archivebox/templates/core/snapshot.html
+++ b/archivebox/templates/core/snapshot.html
@@ -113,6 +113,10 @@
                border-radius: 10px;
                background-color: black;
                overflow: hidden;
+                min-height: 130px;
+            }
+            .header-bottom-frames .card:has([data-compact]) {
+                min-height: 0;
            }
            .card h4 {
                font-size: 1.4vw;
@@ -154,6 +158,202 @@
                transform: scale(0.25); 
                transform-origin: 0 0;
            }
+            .row.header-bottom-frames {
+                display: block !important;
+                width: 100%;
+                max-width: 100%;
+                column-width: 180px;
+                column-gap: 8px;
+                column-fill: auto;
+                margin-left: 0px;
+                margin-right: 0px;
+                flex: none !important;
+            }
+            .header-bottom-frames .col-lg-2 {
+                padding-left: 0px;
+                padding-right: 0px;
+                max-width: 100%;
+                width: 100% !important;
+                display: inline-block !important;
+                float: none !important;
+                flex: none !important;
+                break-inside: avoid;
+                margin-bottom: 6px;
+                vertical-align: top;
+            }
+            .header-bottom-frames .card:has([data-compact]) .thumbnail-wrapper,
+            .header-bottom-frames .card:has([data-compact]) .thumbnail-wrapper.compact {
+                height: 32px;
+            }
+            .header-bottom-frames .card:has([data-compact]) .thumbnail-text {
+                height: auto;
+                max-height: 64px;
+            }
+            .header-bottom-frames .card:has([data-compact]) .card-body {
+                padding: 4px 8px;
+                max-height: 44px;
+            }
+            .thumbnail-wrapper {
+                height: 100px;
+                overflow: hidden;
+                background: #333;
+            }
+            .thumbnail-compact {
+                height: 32px;
+                display: flex;
+                align-items: center;
+                gap: 6px;
+                padding: 0 8px;
+                font-size: 13px;
+                line-height: 1;
+                color: #bdbdbd;
+                background: #111;
+                border-bottom: 1px solid #222;
+                text-transform: uppercase;
+                letter-spacing: 0.02em;
+            }
+            .thumbnail-compact-label {
+                color: #e1e1e1;
+            }
+            .thumbnail-compact-meta {
+                color: #777;
+                font-size: 11px;
+                margin-left: auto;
+            }
+            .thumbnail-compact svg,
+            .thumbnail-compact img {
+                height: 12px;
+                width: 12px;
+            }
+            .thumbnail-text {
+                height: 100px;
+                background: #121212;
+                color: #d8d8d8;
+                padding: 6px 8px;
+                display: flex;
+                flex-direction: column;
+                gap: 4px;
+                font-family: SFMono-Regular, Menlo, Monaco, Consolas, "Liberation Mono", "Courier New", monospace;
+                font-size: 11px;
+                line-height: 1.2;
+                overflow: hidden;
+            }
+            .thumbnail-text-header {
+                display: flex;
+                align-items: center;
+                gap: 6px;
+                font-size: 10px;
+                color: #9b9b9b;
+                text-transform: uppercase;
+                letter-spacing: 0.04em;
+            }
+            .thumbnail-text-pre {
+                margin: 0;
+                white-space: pre-wrap;
+                overflow: hidden;
+                display: -webkit-box;
+                -webkit-box-orient: vertical;
+                -webkit-line-clamp: 5;
+            }
+            .thumbnail-text[data-plugin="title"] .thumbnail-text-pre {
+                font-size: 13px;
+                font-weight: 600;
+                -webkit-line-clamp: 3;
+            }
+            .thumb-grid {
+                display: block;
+                column-width: 180px;
+                column-gap: 6px;
+                align-content: start;
+                width: 100%;
+            }
+            .thumb-card {
+                box-shadow: 2px 3px 14px 0px rgba(0,0,0,0.02);
+                border: 1px solid rgba(0,0,0,3);
+                border-radius: 10px;
+                background-color: black;
+                overflow: hidden;
+                display: inline-block;
+                width: 100%;
+                break-inside: avoid;
+                box-sizing: border-box;
+                margin-bottom: 6px;
+                height: 138px;
+                min-height: 138px;
+                max-height: 138px;
+            }
+            .thumb-card:has([data-compact]) {
+                height: 46px;
+                min-height: 46px;
+                max-height: 46px;
+            }
+            .thumb-card .thumb-body {
+                font-size: 14px;
+                padding: 3px 8px;
+                line-height: 1.2;
+                word-wrap: break-word;
+                overflow: hidden;
+                text-overflow: ellipsis;
+                background-color: #1a1a1a;
+                color: #d3d3d3;
+            }
+            .thumb-card .thumb-body h4 {
+                font-size: 1.1em;
+                margin: 0 0 2px 0;
+                line-height: 1.1;
+                overflow: hidden;
+                text-overflow: ellipsis;
+                white-space: nowrap;
+            }
+            .thumb-card .thumbnail-wrapper,
+            .thumb-card iframe.card-img-top {
+                display: block;
+                width: 100%;
+            }
+            .thumb-card:has([data-compact]) .thumbnail-wrapper,
+            .thumb-card:has([data-compact]) .thumbnail-wrapper.compact {
+                height: 24px;
+                flex: 0 0 auto;
+            }
+            .thumb-card:has([data-compact]) .thumb-body {
+                padding: 2px 6px;
+                font-size: 12px;
+                max-height: 20px;
+            }
+            .thumb-card:has([data-compact]) .thumb-body h4 {
+                font-size: 0.9em;
+                margin-bottom: 0px;
+                line-height: 1;
+                display: flex;
+                align-items: center;
+                gap: 4px;
+                overflow: hidden;
+                text-overflow: ellipsis;
+                white-space: nowrap;
+            }
+            .thumb-card.selected-card {
+                border: 2px solid orange;
+                box-shadow: 0px -6px 13px 1px rgba(0,0,0,0.05);
+            }
+            .thumb-compact .thumbnail-wrapper {
+                height: 32px;
+            }
+            .thumb-compact {
+                margin-bottom: 0px;
+                border-radius: 6px;
+            }
+            .thumb-compact .card-body {
+                display: block;
+                padding: 4px 8px;
+                font-size: 12px;
+                line-height: 1.2;
+                max-height: none;
+            }
+            .thumb-compact .thumbnail-compact,
+            .thumb-compact .thumbnail-text {
+                height: 32px;
+                max-height: 32px;
+            }
            .full-page-iframe {
                border-top: 1px solid #ddd;
                width: 100%;
@@ -203,6 +403,10 @@
                box-shadow: 4px 4px 4px rgba(0,0,0,0.2);
                margin-top: 0px;
            }
+            .header-bottom.container-fluid {
+                padding-left: 6px;
+                padding-right: 6px;
+            }
            .header-bottom-info {
                color: #6f6f6f;
                padding-top: 0px;
@@ -357,15 +561,15 @@
                        </div>
                    </div>
                </div>
-                <div class="row header-bottom-frames">
+                <div class="thumb-grid">
                    {% for result_info in archiveresults %}
                        {% if result_info.result %}
-                            {% plugin_thumbnail result_info.result as thumbnail_html %}
-                            {% if thumbnail_html %}
-                            <div class="col-lg-2">
-                                <div class="card{% if forloop.first %} selected-card{% endif %}">
-                                    {{ thumbnail_html }}
-                                    <div class="card-body">
+                            {% plugin_card result_info.result as thumbnail_html %}
+                            <div class="thumb-card{% if forloop.first %} selected-card{% endif %}">
+                                    <div class="thumbnail-wrapper">
+                                        {{ thumbnail_html }}
+                                    </div>
+                                    <div class="thumb-body">
                                        <a href="{{ result_info.path }}" title="Open in new tab..." target="_blank" rel="noopener">
                                            <p class="card-text"><code>{{ result_info.path }}</code></p>
                                        </a>
@@ -373,18 +577,15 @@
                                            <h4 class="card-title">{{ result_info.name|title }}</h4>
                                        </a>
                                    </div>
-                                </div>
                            </div>
-                            {% endif %}
                        {% endif %}
                    {% endfor %}

                    {% get_config "PREVIEW_ORIGINALS" as preview_originals %}
                    {% if preview_originals %}
-                    <div class="col-lg-2">
-                        <div class="card">
+                    <div class="thumb-card">
                            <iframe class="card-img-top" src="{{url}}" sandbox="allow-same-origin allow-top-navigation-by-user-activation allow-scripts allow-forms" scrolling="no" loading="lazy" referrerpolicy="no-referrer"></iframe>
-                            <div class="card-body">
+                            <div class="thumb-body">
                                <a href="{{url}}" title="Open in new tab..." target="_blank" rel="noopener" referrerpolicy="no-referrer">
                                    <p class="card-text"><code>🌐 {{domain}}</code></p>
                                </a>
@@ -392,7 +593,6 @@
                                    <h4 class="card-title">Original</h4>
                                </a>
                            </div>
-                        </div>
                    </div>
                    {% endif %}
                </div>
@@ -417,19 +617,31 @@
            }

            // show selected file in iframe when preview card is clicked
-            jQuery('.card').on('click', function(e) {
+            jQuery('.thumb-card').on('click', function(e) {
                jQuery('.selected-card').removeClass('selected-card')
-                jQuery(e.target).closest('.card').addClass('selected-card')
-            })
-            jQuery('.card a[target=preview]').on('click', function(e) {
-                if (e.currentTarget.href.endsWith('.pdf')) {
-                    jQuery('.full-page-iframe')[0].removeAttribute('sandbox')
-                } else {
-                    jQuery('.full-page-iframe')[0].sandbox = "allow-same-origin allow-scripts allow-forms allow-top-navigation-by-user-activation"
+                jQuery(e.target).closest('.thumb-card').addClass('selected-card')
+
+                const link = e.target.closest('a[target=preview]') || e.currentTarget.querySelector('a[target=preview]') || e.currentTarget.querySelector('a')
+                if (!link || !link.href || link.href.endsWith('#')) {
+                    return true
                }
-                window.location.hash = getPreviewTypeFromPath(e.currentTarget)
+                const iframe = jQuery('.full-page-iframe')[0]
+                if (!iframe) {
+                    return true
+                }
+                if (link.href.endsWith('.pdf')) {
+                    iframe.removeAttribute('sandbox')
+                } else {
+                    iframe.sandbox = "allow-same-origin allow-scripts allow-forms allow-top-navigation-by-user-activation"
+                }
+                window.location.hash = getPreviewTypeFromPath(link)
+                iframe.src = link.href
                return true
            })
+            jQuery('.thumb-card a[target=preview]').on('click', function(e) {
+                e.preventDefault()
+                return false
+            })

            function hideSnapshotHeader() {
                console.log('Collapsing Snapshot header...')
@@ -483,7 +695,7 @@
                    for (const link of jQuery('a[target=preview]')) {
                        console.log(link.pathname)
                        if (getPreviewTypeFromPath(link) == window.location.hash.slice(1).toLowerCase()) {
-                            jQuery(link).closest('.card').click()
+                            jQuery(link).closest('.thumb-card').click()
                            jQuery(link).click()
                            link.click()
                        }
@@ -502,7 +714,7 @@

            // hide all preview iframes on small screens
            if (window.innerWidth < 1091) {
-                jQuery('.card a[target=preview]').attr('target', '_self')
+            jQuery('.thumb-card a[target=preview]').attr('target', '_self')
            }

            var pdf_frame = document.querySelector('.pdf-frame');
--- a/archivebox/templates/core/snapshot_live.html
+++ b/archivebox/templates/core/snapshot_live.html
@@ -130,12 +130,15 @@

            .header-bottom-frames .card {
                box-shadow: 2px 2px 7px 0px rgba(0, 0, 0, 0.1);
-                margin-bottom: 5px;
+                margin-bottom: 6px;
                border: 1px solid rgba(0, 0, 0, 0.06);
                border-radius: 10px;
                background-color: #efefef;
                overflow: hidden;
-                height: 130px;
+                min-height: 130px;
+            }
+            .header-bottom-frames .card:has([data-compact]) {
+                min-height: 0;
            }
            .card h4 {
                font-size: 0.8em;
@@ -144,7 +147,7 @@
                text-transform: uppercase;
                margin-top: 0px;
                margin-bottom: 5px;
-                color: rgb(93, 105, 110);
+                color: #222;
            }
            .card-body {
                font-size: 14px;
@@ -158,7 +161,8 @@
                max-height: 102px;
                overflow: hidden;
                text-overflow: ellipsis;
-                color: #d3d3d3;
+                color: #222;
+                background-color: #f6f6f6;
            }
            .card-title {
                margin-bottom: 4px;
@@ -213,6 +217,10 @@
                background-color: #333;
                pointer-events: none;
            }
+            .thumbnail-wrapper.compact {
+                height: 32px;
+                background-color: #111;
+            }
            .thumbnail-wrapper iframe {
                width: 405%;
                height: 430px;
@@ -228,10 +236,89 @@
                object-fit: cover;
                object-position: top center;
            }
+            .thumbnail-compact {
+                height: 32px;
+                display: flex;
+                align-items: center;
+                gap: 6px;
+                padding: 0 8px;
+                font-size: 13px;
+                line-height: 1;
+                color: #bdbdbd;
+                text-transform: uppercase;
+                letter-spacing: 0.02em;
+            }
+            .thumbnail-compact-label {
+                color: #e1e1e1;
+            }
+            .thumbnail-compact-meta {
+                color: #777;
+                font-size: 11px;
+                margin-left: auto;
+            }
+            .thumbnail-compact svg,
+            .thumbnail-compact img {
+                height: 12px;
+                width: 12px;
+            }
+            .thumbnail-text {
+                height: 100px;
+                background: #121212;
+                color: #d8d8d8;
+                padding: 6px 8px;
+                display: flex;
+                flex-direction: column;
+                gap: 4px;
+                font-family: SFMono-Regular, Menlo, Monaco, Consolas, "Liberation Mono", "Courier New", monospace;
+                font-size: 11px;
+                line-height: 1.2;
+                overflow: hidden;
+            }
+            .thumbnail-text-header {
+                display: flex;
+                align-items: center;
+                gap: 6px;
+                font-size: 10px;
+                color: #9b9b9b;
+                text-transform: uppercase;
+                letter-spacing: 0.04em;
+            }
+            .thumbnail-text-pre {
+                margin: 0;
+                white-space: pre-wrap;
+                overflow: hidden;
+                display: -webkit-box;
+                -webkit-box-orient: vertical;
+                -webkit-line-clamp: 5;
+            }
+            .thumbnail-text[data-plugin="title"] .thumbnail-text-pre {
+                font-size: 13px;
+                font-weight: 600;
+                -webkit-line-clamp: 3;
+            }
            .card.selected-card {
                border: 2px solid orange;
                box-shadow: 0px -6px 13px 1px rgba(0,0,0,0.05);
            }
+            .thumb-compact .card-body {
+                display: block;
+                padding: 4px 8px;
+                font-size: 12px;
+                line-height: 1.2;
+                max-height: none;
+            }
+            .thumb-compact {
+                margin-bottom: 0px;
+                border-radius: 6px;
+            }
+            .thumb-compact .thumbnail-wrapper {
+                height: 32px;
+            }
+            .thumb-compact .thumbnail-compact,
+            .thumb-compact .thumbnail-text {
+                height: 32px;
+                max-height: 32px;
+            }
            .iframe-large {
                height: calc(100vh - 70px);
            }
@@ -256,6 +343,83 @@
                object-fit: cover;
                object-position: top center;
            }
+            .thumb-grid {
+                display: block;
+                column-width: 180px;
+                column-gap: 6px;
+                align-content: start;
+                width: 100%;
+            }
+            .thumb-card {
+                box-shadow: 2px 2px 7px 0px rgba(0, 0, 0, 0.1);
+                border: 1px solid rgba(0, 0, 0, 0.06);
+                border-radius: 10px;
+                background-color: #efefef;
+                overflow: hidden;
+                display: inline-block;
+                width: 100%;
+                break-inside: avoid;
+                box-sizing: border-box;
+                margin-bottom: 6px;
+                height: 138px;
+                min-height: 138px;
+                max-height: 138px;
+            }
+            .thumb-card:has([data-compact]) {
+                height: 46px;
+                min-height: 46px;
+                max-height: 46px;
+            }
+            .thumb-card .thumb-body {
+                font-size: 14px;
+                padding: 3px 8px;
+                line-height: 1.2;
+                word-wrap: break-word;
+                overflow: hidden;
+                text-overflow: ellipsis;
+                color: #222;
+                background-color: #f6f6f6;
+            }
+            .thumb-card .thumb-body h4 {
+                font-size: 0.8em;
+                text-transform: uppercase;
+                margin: 0 0 2px 0;
+                color: #222;
+                line-height: 1.1;
+                overflow: hidden;
+                text-overflow: ellipsis;
+                white-space: nowrap;
+            }
+            .thumb-card .thumbnail-wrapper,
+            .thumb-card iframe.card-img-top {
+                display: block;
+                width: 100%;
+            }
+            .thumb-card:has([data-compact]) .thumbnail-wrapper,
+            .thumb-card:has([data-compact]) .thumbnail-wrapper.compact {
+                height: 24px;
+                flex: 0 0 auto;
+            }
+            .thumb-card:has([data-compact]) .thumb-body {
+                padding: 2px 6px;
+                font-size: 12px;
+                max-height: 20px;
+            }
+            .thumb-card:has([data-compact]) .thumb-body h4 {
+                font-size: 0.9em;
+                margin-bottom: 0px;
+                line-height: 1;
+                display: flex;
+                align-items: center;
+                gap: 4px;
+                overflow: hidden;
+                text-overflow: ellipsis;
+                white-space: nowrap;
+            }
+            .thumb-card.selected-card {
+                border: 2px solid orange;
+                box-shadow: 0px -6px 13px 1px rgba(0,0,0,0.05);
+            }
            .header-bottom {
                border-top: 1px solid rgba(170, 30, 85, 0.9);
                padding-bottom: 1px;
@@ -268,6 +432,10 @@
                box-shadow: 4px 4px 4px rgba(0,0,0,0.2);
                margin-top: 0px;
            }
+            .header-bottom.container-fluid {
+                padding-left: 6px;
+                padding-right: 6px;
+            }
            .header-bottom-info {
                color: #6f6f6f;
                padding-top: 0px;
@@ -315,9 +483,41 @@
                width: 100%;
                overflow: hidden;
            }
-            .header-bottom-frames {
+            .row.header-bottom-frames {
                padding-top: 5px;
-                justify-content: center;
+                display: block !important;
+                width: 100%;
+                max-width: 100%;
+                column-width: 180px;
+                column-gap: 8px;
+                column-fill: auto;
+                margin-left: 0px;
+                margin-right: 0px;
+                flex: none !important;
+            }
+            .header-bottom-frames .col-lg-2 {
+                padding-left: 0px;
+                padding-right: 0px;
+                max-width: 100%;
+                width: 100% !important;
+                display: inline-block !important;
+                float: none !important;
+                flex: none !important;
+                break-inside: avoid;
+                margin-bottom: 6px;
+                vertical-align: top;
+            }
+            .header-bottom-frames .card:has([data-compact]) .thumbnail-wrapper,
+            .header-bottom-frames .card:has([data-compact]) .thumbnail-wrapper.compact {
+                height: 32px;
+            }
+            .header-bottom-frames .card:has([data-compact]) .thumbnail-text {
+                height: auto;
+                max-height: 64px;
+            }
+            .header-bottom-frames .card:has([data-compact]) .card-body {
+                padding: 4px 8px;
+                max-height: 44px;
            }
            .header-bottom-frames .card-title {
                width: 100%;
@@ -325,7 +525,7 @@
                font-size: 17px;
                margin-bottom: 0px;
                display: inline-block;
-                color: #d3d3d3;
+                color: #222;
                font-weight: 200;
                vertical-align: 3px;
            }
@@ -415,7 +615,7 @@
                        </small>
                    </div>
                    <div class="col-lg-2" style="padding-top: 4px">
-                        <a href="/archive/{{url}}" title="Date Added: {{bookmarked_date}}  |  First Archived: {{oldest_archive_date|default:downloaded_datestr}}  |  Last Checked: {{downloaded_datestr}}   (UTC)">
+                        <a href="/{{archive_path}}/index.html" title="Date Added: {{bookmarked_date}}  |  First Archived: {{oldest_archive_date|default:downloaded_datestr}}  |  Last Checked: {{downloaded_datestr}}   (UTC)">
                            {{oldest_archive_date|default:downloaded_datestr|default:bookmarked_date}}
                        </a>
                        <br/>
@@ -431,34 +631,45 @@
                </div>
            </div>
            <div class="header-bottom container-fluid">
-                <div class="row header-bottom-frames">
+                <div class="thumb-grid">
                    
                    
                    {% for result in archiveresults %}
-                        <div class="col-lg-2">
-                            <div class="card {% if forloop.first %}selected-card{% endif %}">
-                                <div class="card-body">
-                                    <a href="{{result.path|urlencode}}" target="preview" title="./{{result.path}} (downloaded {{result.ts}})">
+                        <div class="thumb-card{% if forloop.first %} selected-card{% endif %}">
+                            {% with display_path=result.path %}
+                                <div class="thumb-body">
+                                    {% if display_path %}
+                                        <a href="{{display_path|urlencode}}" target="preview" title="./{{display_path}} (downloaded {{result.ts}})">
+                                            <h4>{% plugin_icon result.name %} {{result.name|plugin_name|truncatechars:20}} <small>({{result.size|filesizeformat}})</small></h4>
+                                        </a>
+                                    {% else %}
                                        <h4>{% plugin_icon result.name %} {{result.name|plugin_name|truncatechars:20}} <small>({{result.size|filesizeformat}})</small></h4>
-                                    </a>
+                                    {% endif %}
                                </div>
-                                {% if result.result %}
-                                    {# Use plugin-specific thumbnail template when ArchiveResult is available #}
+                                {% if result.result and display_path %}
+                                    {# Use plugin-specific card template when ArchiveResult is available #}
                                    <div class="card-img-top thumbnail-wrapper">
-                                        {% plugin_thumbnail result.result %}
+                                        {% plugin_card result.result %}
                                    </div>
-                                {% else %}
+                                {% elif result.is_metadata and display_path %}
+                                    <div class="card-img-top thumbnail-wrapper compact">
+                                        <div class="thumbnail-compact" data-plugin="{{result.name}}">
+                                            <span class="thumbnail-compact-icon">{% plugin_icon result.name %}</span>
+                                            <span class="thumbnail-compact-label">{{result.name|plugin_name}}</span>
+                                            <span class="thumbnail-compact-meta">metadata</span>
+                                        </div>
+                                    </div>
+                                {% elif display_path %}
                                    {# Fall back to generic iframe for filesystem-discovered files #}
-                                    <iframe class="card-img-top" src="{{result.path|urlencode}}?autoplay=0" allow="autoplay 'none'; fullscreen 'none'; navigation-override 'none'; " sandbox="allow-same-origin allow-top-navigation-by-user-activation allow-scripts allow-forms" scrolling="no" loading="lazy"></iframe>
+                                    <iframe class="card-img-top" src="{{display_path|urlencode}}?autoplay=0" allow="autoplay 'none'; fullscreen 'none'; navigation-override 'none'; " sandbox="allow-same-origin allow-top-navigation-by-user-activation allow-scripts allow-forms" scrolling="no" loading="lazy"></iframe>
                                {% endif %}
-                            </div>
+                            {% endwith %}
                        </div>
                    {% endfor %}


-                    <div class="col-lg-2">
-                        <div class="card">
-                            <div class="card-body">
+                    <div class="thumb-card">
+                            <div class="thumb-body">
                                <a href="./" target="preview">
                                    <h4>Headers, JSON, etc.</h4>
                                </a>
@@ -466,7 +677,6 @@
                            </div>
                            <iframe class="card-img-top" src="./" sandbox="" scrolling="no" loading="lazy"></iframe>
                        </div>
-                    </div>
                </div>
            </div>
        </header>
@@ -476,11 +686,11 @@
        {% if best_result.result %}
            {# Use plugin-specific fullscreen template when ArchiveResult is available #}
            <div id="main-frame-wrapper" class="full-page-wrapper">
-                {% plugin_fullscreen best_result.result %}
+                {% plugin_full best_result.result %}
            </div>
        {% else %}
            {# Fall back to generic iframe #}
-            <iframe id="main-frame" sandbox="allow-same-origin allow-top-navigation-by-user-activation allow-scripts allow-forms" class="full-page-iframe" src="{{best_result.path|urlencode}}" name="preview"></iframe>
+            <iframe id="main-frame" sandbox="allow-same-origin allow-top-navigation-by-user-activation allow-scripts allow-forms" class="full-page-iframe" src="{{best_result.path|default:'about:blank'|urlencode}}" name="preview"></iframe>
        {% endif %}
    

@@ -513,21 +723,45 @@
                return link.getAttribute('href')
            }

-            const iframe_elem = document.getElementById('main-frame')
+            function ensureMainFrame() {
+                let frame = document.getElementById('main-frame')
+                if (!frame) {
+                    const wrapper = document.getElementById('main-frame-wrapper')
+                    frame = document.createElement('iframe')
+                    frame.id = 'main-frame'
+                    frame.name = 'preview'
+                    frame.className = 'full-page-iframe'
+                    frame.sandbox = "allow-same-origin allow-top-navigation-by-user-activation allow-scripts allow-forms"
+                    if (wrapper) {
+                        wrapper.innerHTML = ''
+                        wrapper.appendChild(frame)
+                        wrapper.classList.remove('full-page-wrapper')
+                    }
+                }
+                return frame
+            }

-            for (const card of [...document.querySelectorAll('.card')]) {
+            for (const card of [...document.querySelectorAll('.thumb-card')]) {
                card.addEventListener('click', function(event) {
-                    const target = event.currentTarget.querySelector('a').href
+                    const link = event.target.closest('a[target=preview]') || event.currentTarget.querySelector('a[target=preview]') || event.currentTarget.querySelector('a')
+                    if (!link) {
+                        return
+                    }
+                    const target = link.href
+                    if (!target || target.endsWith('#')) {
+                        return
+                    }

                    jQuery('.selected-card').removeClass('selected-card')
-                    jQuery(event.currentTarget).closest('.card').addClass('selected-card')
+                    jQuery(event.currentTarget).closest('.thumb-card').addClass('selected-card')

+                    const iframe_elem = ensureMainFrame()
                    if (target.endsWith('.pdf')) {
-                        jQuery('#main-frame')[0].removeAttribute('sandbox')
+                        iframe_elem.removeAttribute('sandbox')
                    } else {
-                        jQuery('#main-frame')[0].sandbox = "allow-same-origin allow-top-navigation-by-user-activation allow-scripts allow-forms"
+                        iframe_elem.sandbox = "allow-same-origin allow-top-navigation-by-user-activation allow-scripts allow-forms"
                    }
-                    window.location.hash = getPreviewTypeFromPath(event.currentTarget.querySelector('a'))
+                    window.location.hash = getPreviewTypeFromPath(link)

                    iframe_elem.src = target
                })
@@ -587,7 +821,7 @@
                    for (const link of jQuery('a[target=preview]')) {
                        console.log(link.pathname)
                        if (getPreviewTypeFromPath(link) == window.location.hash.slice(1).toLowerCase()) {
-                            jQuery(link).closest('.card').click()
+                            jQuery(link).closest('.thumb-card').click()
                            jQuery(link).click()
                            link.click()
                        }
--- a/archivebox/workers/worker.py
+++ b/archivebox/workers/worker.py
@@ -698,7 +698,7 @@ class SnapshotWorker(Worker):

        try:
            # Get merged config (includes env vars passed via Process.env, snapshot.config, defaults, etc.)
-            config = get_config(snapshot=self.snapshot)
+            config = get_config(snapshot=self.snapshot, crawl=self.snapshot.crawl)

            # Discover all hooks for this snapshot
            hooks = discover_hooks('Snapshot', config=config)
@@ -842,14 +842,13 @@ class SnapshotWorker(Worker):
        # Clear to avoid double-termination during on_shutdown
        self.background_processes = {}

-        # Update STARTED background results now that hooks are done
+        # Update background results now that hooks are done
        from archivebox.core.models import ArchiveResult

-        started_bg = self.snapshot.archiveresult_set.filter(
-            status=ArchiveResult.StatusChoices.STARTED,
+        bg_results = self.snapshot.archiveresult_set.filter(
            hook_name__contains='.bg.',
        )
-        for ar in started_bg:
+        for ar in bg_results:
            ar.update_from_output()

    def _reap_background_hooks(self) -> None:
@@ -867,7 +866,7 @@ class SnapshotWorker(Worker):
                continue

            ar = self.snapshot.archiveresult_set.filter(hook_name=hook_name).first()
-            if ar and ar.status == ArchiveResult.StatusChoices.STARTED:
+            if ar:
                ar.update_from_output()

            # Remove completed hook from tracking