KeyHunter parser v10 + 18 scanners + tests (synthetic fixtures only, no keys/db/tokens)

merge_pat.py

@@ -0,0 +1,81 @@
+"""Merge PATs from ACCOUNTS_PAT.txt (and other pools) into gh_tokens.txt.
+Live-verify each PAT via /user endpoint; dedup; append only new valid ones."""
+import os
+KF_BASE = os.path.dirname(os.path.abspath(__file__))
+import json
+import time
+import urllib.request
+
+POOLS = [
+    r"C:/Users/User/tmp/ACCOUNTS_PAT.txt",
+]
+TARGET = os.path.join(KF_BASE, "gh_tokens.txt")
+UA = {"User-Agent": "Mozilla/5.0", "Accept": "application/vnd.github+json"}
+
+
+def log(msg):
+    print(f"[{time.strftime('%H:%M:%S')}] {msg}", flush=True)
+
+
+def check(pat):
+    try:
+        req = urllib.request.Request("https://api.github.com/user", headers={**UA, "Authorization": f"token {pat}"})
+        with urllib.request.urlopen(req, timeout=20) as r:
+            if r.status == 200:
+                d = json.loads(r.read())
+                return d.get("login", "?")
+    except Exception as e:
+        return None
+    return None
+
+
+def load_json_pool(path):
+    pats = []
+    try:
+        d = json.load(open(path, encoding="utf-8"))
+        for a in d:
+            if isinstance(a, dict) and a.get("github_pat"):
+                pats.append(a["github_pat"])
+    except Exception:
+        pass
+    return pats
+
+
+def main():
+    with open(TARGET, encoding="utf-8") as f:
+        existing = {line.strip() for line in f if line.strip().startswith("ghp_")}
+    log(f"existing tokens: {len(existing)}")
+    total_new = 0
+    for pool in POOLS:
+        try:
+            with open(pool, encoding="utf-8", errors="replace") as f:
+                lines = [l.strip() for l in f if l.strip() and not l.startswith("#")]
+        except FileNotFoundError:
+            log(f"pool {pool} not found, skip")
+            continue
+        pats = []
+        for l in lines:
+            # login | pat | totp | password | email
+            parts = [p.strip() for p in l.split("|")]
+            for p in parts:
+                if p.startswith("ghp_") or p.startswith("github_pat_"):
+                    pats.append(p)
+        log(f"{pool}: {len(pats)} PATs parsed")
+        for i, pat in enumerate(pats):
+            if pat in existing:
+                continue
+            login = check(pat)
+            if login:
+                existing.add(pat)
+                total_new += 1
+                with open(TARGET, "a", encoding="utf-8") as f:
+                    f.write(pat + "\n")
+                log(f"  + {login} {pat[:14]}... valid, appended")
+            else:
+                log(f"  x {pat[:14]}... dead")
+            time.sleep(0.4)
+    log(f"DONE: +{total_new} new live PATs, total {len(existing)}")
+
+
+if __name__ == "__main__":
+    main()