Feat: Enhance library management features

- Adds strict HTTP/HTTPS validation when creating library websites.
- Improves summary cleanup in video ingestion by removing thinking tags and prefixes.
- Updates LibraryManager UI to display comprehensive metadata (ID, provider, upload date, character count, etc.) and improves video source handling.

backend/local_rag.py

@@ -14,7 +14,7 @@ from concurrent.futures import ThreadPoolExecutor
 from datetime import datetime
 from pathlib import Path
 from typing import Any, Dict, List, Optional
-from urllib.parse import quote
+from urllib.parse import quote, urlparse
 
 import httpx
 from fastapi import APIRouter, HTTPException
@@ -1599,6 +1599,9 @@ async def update_library_text(slug: str, item_id: str, req: UpdateTextRequest):
 @router.post("/libraries/{slug}/websites")
 async def create_library_website(slug: str, req: CreateWebsiteRequest):
     requested_url = str(req.url or "").strip()
+    parsed_url = urlparse(requested_url)
+    if parsed_url.scheme not in {"http", "https"} or not parsed_url.netloc:
+        raise HTTPException(status_code=400, detail="Enter a valid HTTP or HTTPS URL.")
     try:
         video_metadata = await asyncio.to_thread(probe_video_url, requested_url)
     except UnsupportedVideoUrl:

backend/video_ingest.py

@@ -351,7 +351,8 @@ async def _summarize(title: str, transcript: str, model_name: str) -> str:
         ],
         options={"num_ctx": _choose_num_ctx(prompt)},
     )
-    clean = str(summary or "").strip()
+    clean = re.sub(r"<think>.*?</think>", "", str(summary or ""), flags=re.DOTALL | re.IGNORECASE).strip()
+    clean = re.sub(r"^\s*Summary:\s*", "", clean, flags=re.IGNORECASE).strip()
     if not clean:
         raise RuntimeError("Ollama returned an empty video summary.")
     return clean