Align trending PoCs and filter to recent data

scripts/build_joined.py

@@ -24,6 +24,7 @@ EPSS_PATH = DATA_DIR / "epss.json"
 DEFAULT_TOP_KEV = 75
 DEFAULT_HIGH_EPSS_LIMIT = 50
 DEFAULT_HIGH_EPSS_THRESHOLD = 0.05
+RECENT_YEAR_WINDOW = 1
 
 
 def load_inputs(kev_path: Path, epss_path: Path) -> Tuple[Dict, Dict]:
@@ -34,11 +35,22 @@ def load_inputs(kev_path: Path, epss_path: Path) -> Tuple[Dict, Dict]:
 
 def enrich_kev(kev_items: List[Dict], epss_lookup: Dict[str, Dict], poc_index: Dict[str, Dict]) -> List[Dict]:
     enriched = []
+    current_year = today_str()
+    current_year_int = int(current_year.split("-")[0])
+    def is_recent(cve_id: str) -> bool:
+        try:
+            year = int(cve_id.split("-")[1])
+        except Exception:
+            return False
+        return year >= current_year_int - RECENT_YEAR_WINDOW
+
     for entry in kev_items:
         cve = entry.get("cve") or entry.get("cveID") or ""
         if not cve:
             continue
         cve = cve.upper()
+        if not is_recent(cve):
+            continue
         epss_info = epss_lookup.get(cve, {})
         poc_info = poc_index.get(cve)
         if not poc_info or not poc_info.get("poc"):
@@ -75,6 +87,14 @@ def build_high_epss_not_in_kev(
     threshold: float,
     limit: int,
 ) -> List[Dict]:
+    current_year_int = int(today_str().split("-")[0])
+    def is_recent(cve_id: str) -> bool:
+        try:
+            year = int(cve_id.split("-")[1])
+        except Exception:
+            return False
+        return year >= current_year_int - RECENT_YEAR_WINDOW
+
     ranked = sorted(
         (
             row
@@ -82,6 +102,7 @@ def build_high_epss_not_in_kev(
             if row.get("cve")
             and row.get("cve", "").upper() not in kev_set
             and (row.get("epss") is not None)
+            and is_recent(row.get("cve", ""))
         ),
         key=lambda row: (-float(row.get("epss") or 0), row.get("cve", "")),
     )

scripts/build_site.py

@@ -1,7 +1,9 @@
 from __future__ import annotations
 
 import argparse
+from datetime import datetime, timezone
 from pathlib import Path
+import re
 from typing import Dict, Tuple
 
 from jinja2 import Environment, FileSystemLoader, select_autoescape
@@ -59,7 +61,29 @@ def build_pages(env: Environment, data: Dict, diff: Dict | None = None, html_mod
     joined = data["joined"]
     details = data["details"]
     vendors = data["vendors"]
-    trending = parse_trending_from_readme(README_PATH)
+    def is_recent_label(label: str) -> bool:
+        label = (label or "").lower()
+        if "minute" in label or "hour" in label:
+            return True
+        m = re.search(r"(\d+)\\s*day", label)
+        if not m:
+            return False
+        return int(m.group(1)) <= 4
+
+    current_year = datetime.now(timezone.utc).year
+
+    def extract_year(name: str) -> int | None:
+        m = re.search(r"cve-(\\d{4})-", name.lower())
+        return int(m.group(1)) if m else None
+
+    trending_raw = parse_trending_from_readme(README_PATH)
+    trending = [
+        row
+        for row in trending_raw
+        if is_recent_label(row.get("updated", ""))
+        and (extract_year(row.get("name", "")) or current_year) >= current_year - 1
+    ]
+    trending.sort(key=lambda r: int(r.get("stars") or 0), reverse=True)
     recent_kev = (diff or {}).get("new_kev_entries") or []
     metrics = {
         "kev_total": len(data["kev_enriched"]),