add

requirements.txt

@@ -10,6 +10,7 @@ matplotlib
 numpy
 pandas
 python-dateutil
+requests
 tqdm
 transformers
 tokenizers>=0.15.0

src/envs.py

@@ -11,6 +11,10 @@ LOCAL_MODE = True
 # Get token from environment or use None in local mode
 TOKEN = os.environ.get("HF_TOKEN") if not LOCAL_MODE else None
 
+# GitHub API token for fetching repo metadata
+# This increases rate limits from 60 to 5000 requests per hour
+GITHUB_TOKEN = os.environ.get("GITHUB_TOKEN", "")
+
 OWNER = "libvulnwatch" # Change to your org - don't forget to create a results and request dataset, with the correct format!
 # ----------------------------------
 

src/leaderboard/github_data.py

@@ -0,0 +1,131 @@
+"""Utilities for fetching GitHub repository data"""
+
+import os
+import requests
+import time
+from functools import lru_cache
+from urllib.parse import urlparse
+
+# Import GitHub token from envs
+from src.envs import GITHUB_TOKEN
+
+
+def extract_repo_path(repo_url):
+    """Extract org/repo path from GitHub URL
+    
+    Args:
+        repo_url: GitHub repository URL
+        
+    Returns:
+        Repository path in format "org/repo"
+    """
+    if not repo_url:
+        return None
+        
+    # Handle both URL and org/repo format
+    if repo_url.startswith(("http://", "https://")):
+        parsed = urlparse(repo_url)
+        path = parsed.path.strip("/")
+        
+        # Remove .git suffix if present
+        if path.endswith(".git"):
+            path = path[:-4]
+            
+        return path
+    
+    # Already in org/repo format
+    return repo_url
+
+
+@lru_cache(maxsize=128)
+def get_github_data(repo_path, use_token=True):
+    """Fetch repository data from GitHub API
+    
+    Args:
+        repo_path: Repository path in format "org/repo"
+        use_token: Whether to use GitHub token if available
+        
+    Returns:
+        Dictionary with repository data including stars and license
+    """
+    if not repo_path:
+        return {"github_stars": 0, "license": "Unknown"}
+        
+    api_url = f"https://api.github.com/repos/{repo_path}"
+    headers = {"Accept": "application/vnd.github.v3+json"}
+    
+    # Add token for higher rate limits if available
+    if use_token and GITHUB_TOKEN:
+        headers["Authorization"] = f"token {GITHUB_TOKEN}"
+    
+    try:
+        response = requests.get(api_url, headers=headers)
+        
+        if response.status_code == 200:
+            data = response.json()
+            
+            # Extract relevant fields
+            result = {
+                "github_stars": data.get("stargazers_count", 0),
+                "license": data.get("license", {}).get("spdx_id", "Unknown"),
+                "full_name": data.get("full_name", repo_path),
+                "created_at": data.get("created_at", ""),
+                "updated_at": data.get("updated_at", ""),
+                "language": data.get("language", ""),
+                "forks_count": data.get("forks_count", 0),
+                "default_branch": data.get("default_branch", "main"),
+            }
+            
+            # If license is None or "NOASSERTION", use "Unknown"
+            if not result["license"] or result["license"] == "NOASSERTION":
+                result["license"] = "Unknown"
+                
+            return result
+        else:
+            print(f"GitHub API error for {repo_path}: {response.status_code} - {response.text}")
+            return {"github_stars": 0, "license": "Unknown"}
+            
+    except Exception as e:
+        print(f"Error fetching GitHub data for {repo_path}: {e}")
+        return {"github_stars": 0, "license": "Unknown"}
+
+
+def update_assessment_with_github_data(assessment, force_update=False):
+    """Update assessment with data from GitHub
+    
+    Args:
+        assessment: AssessmentResult object
+        force_update: Whether to force update even if values exist
+        
+    Returns:
+        Updated AssessmentResult object
+    """
+    # Skip if no data is missing or if force_update is False
+    if not force_update and assessment.stars > 0 and assessment.license != "?":
+        return assessment
+    
+    # Try getting repo path from library_name first
+    repo_path = None
+    if assessment.library_name and "/" in assessment.library_name:
+        repo_path = assessment.library_name
+    
+    # Fall back to repository_url if available
+    if not repo_path and hasattr(assessment, 'repository_url') and assessment.repository_url:
+        repo_path = extract_repo_path(assessment.repository_url)
+    
+    # If we still don't have a path, reconstruct from org/repo
+    if not repo_path and assessment.org and assessment.repo:
+        repo_path = f"{assessment.org}/{assessment.repo}"
+    
+    # If we found a valid path, fetch and update
+    if repo_path:
+        github_data = get_github_data(repo_path)
+        
+        # Update if data is missing or force_update is True
+        if force_update or assessment.stars == 0:
+            assessment.stars = github_data["github_stars"]
+            
+        if force_update or assessment.license == "?":
+            assessment.license = github_data["license"]
+    
+    return assessment 

src/leaderboard/read_evals.py

@@ -9,6 +9,7 @@ import numpy as np
 
 from src.display.formatting import make_clickable_library, make_clickable_report
 from src.display.utils import AutoEvalColumn, LibraryType, Tasks, Language, AssessmentStatus
+from src.leaderboard.github_data import update_assessment_with_github_data
 
 
 @dataclass
@@ -30,6 +31,7 @@ class AssessmentResult:
     availability: bool = True
     verified: bool = False
     report_url: str = ""  # URL to detailed assessment report
+    repository_url: str = ""  # GitHub repository URL
 
     @classmethod
     def init_from_json_file(self, json_filepath):
@@ -88,6 +90,7 @@ class AssessmentResult:
             verified=assessment.get("independently_verified", False),
             last_update=last_update,
             report_url=assessment.get("report_url", ""),
+            repository_url=assessment.get("repository_url", ""),
         )
 
     def update_with_request_file(self, requests_path):
@@ -99,8 +102,15 @@ class AssessmentResult:
                 request = json.load(f)
             self.library_type = LibraryType.from_str(request.get("library_type", ""))
             self.stars = request.get("stars", 0)
+            # Add repository URL if not already set
+            if not self.repository_url and "repository_url" in request:
+                self.repository_url = request.get("repository_url", "")
         except Exception:
             print(f"Could not find request file for {self.library_name} version {self.version}")
+        
+        # Try to get GitHub stars and license if missing
+        if self.stars == 0 or self.license == "?":
+            update_assessment_with_github_data(self)
 
     def to_dict(self):
         """Converts the Assessment Result to a dict compatible with our dataframe display"""