diff --git a/src/fabledassistant/services/rss.py b/src/fabledassistant/services/rss.py
index 606159f..6b652e7 100644
--- a/src/fabledassistant/services/rss.py
+++ b/src/fabledassistant/services/rss.py
@@ -14,8 +14,6 @@ from fabledassistant.models.rss_feed import RssFeed, RssItem
 
 logger = logging.getLogger(__name__)
 
-# Safety cap on stored content — effectively unlimited for typical articles
-CONTENT_MAX_CHARS = 50_000
 # Keep only items from the last N days to avoid unbounded growth
 ITEM_MAX_AGE_DAYS = 90
 
@@ -80,7 +78,7 @@ def extract_item(entry) -> dict:
         content = raw_content[0].value
     else:
         content = entry.get("summary", "")
-    content = _html_to_text(content)[:CONTENT_MAX_CHARS]
+    content = _html_to_text(content)
 
     pub = None
     if entry.published_parsed:
@@ -210,7 +208,7 @@ async def fetch_and_cache_feed(feed_id: int, url: str) -> int:
                     async with async_session() as session:
                         item = await session.get(RssItem, item_id)
                         if item:
-                            item.content = full_text[:CONTENT_MAX_CHARS]
+                            item.content = full_text
                             await session.commit()
                             await upsert_rss_item_embedding(
                                 item_id, feed_user_id, item.title or "", item.content
diff --git a/tests/test_rss_service.py b/tests/test_rss_service.py
index 29905b9..b08341a 100644
--- a/tests/test_rss_service.py
+++ b/tests/test_rss_service.py
@@ -16,14 +16,16 @@ def test_extract_item_fields():
     assert item["content"] == "A summary"
 
 
-def test_extract_item_truncates_content():
-    """extract_item() should truncate content to CONTENT_MAX_CHARS."""
-    from fabledassistant.services.rss import extract_item, CONTENT_MAX_CHARS
+def test_extract_item_does_not_truncate_content():
+    """extract_item() should store content without truncation."""
+    from fabledassistant.services.rss import extract_item
+    long_text = "x" * 100_000
     entry = MagicMock()
-    entry.get = lambda k, d="": {"summary": "x" * (CONTENT_MAX_CHARS + 1000), "title": "", "link": "", "id": "g"}.get(k, d)
+    entry.get = lambda k, d="": {"summary": long_text, "title": "", "link": "", "id": "g"}.get(k, d)
+    entry.content = []
     entry.published_parsed = None
     item = extract_item(entry)
-    assert len(item["content"]) == CONTENT_MAX_CHARS
+    assert len(item["content"]) == 100_000
 
 
 def test_extract_item_prefers_content_over_summary():