fix: filter non-integer metadata from GitHub languages API response (#322)

PyGithub's get_languages() returns raw API JSON which in some environments
includes non-integer metadata keys (e.g., "url"), causing a TypeError in
sum(). Now filters to integer values only before calculating percentages.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

tests/test_github_scraper.py

@@ -230,6 +230,31 @@ class TestLanguageDetection(unittest.TestCase):
             self.assertIn("languages", scraper.extracted_data)
             self.assertEqual(scraper.extracted_data["languages"], {})
 
+    def test_extract_languages_filters_non_integer_metadata(self):
+        """Test that non-integer metadata keys (e.g., 'url') are filtered out (#322)"""
+        config = {"repo": "xyflow/xyflow", "name": "xyflow", "github_token": None}
+
+        with patch("skill_seekers.cli.github_scraper.Github"):
+            scraper = self.GitHubScraper(config)
+            scraper.repo = Mock()
+            scraper.repo.get_languages.return_value = {
+                "TypeScript": 707330,
+                "Svelte": 95784,
+                "url": "https://api.github.com/repos/xyflow/xyflow/languages",
+            }
+
+            scraper._extract_languages()
+
+            self.assertIn("languages", scraper.extracted_data)
+            self.assertIn("TypeScript", scraper.extracted_data["languages"])
+            self.assertIn("Svelte", scraper.extracted_data["languages"])
+            self.assertNotIn("url", scraper.extracted_data["languages"])
+
+            # Percentages should be calculated only from real languages
+            ts_data = scraper.extracted_data["languages"]["TypeScript"]
+            total = 707330 + 95784
+            self.assertEqual(ts_data["percentage"], round(707330 / total * 100, 2))
+
 
 class TestIssuesExtraction(unittest.TestCase):
     """Test GitHub Issues extraction (C1.7)"""