""" Stake.com API discovery — strategy: 1. Load the homepage (which passes CF), grab all JS bundle URLs 2. Download and parse bundles for GraphQL query strings 3. Also try the stake.com API with cf_clearance cookie """ import json import time import re from datetime import datetime captured = { "js_bundles": [], "gql_queries_found": [], "gql_operation_names": set(), "cookies": {}, "api_responses": [], } def run(): from playwright.sync_api import sync_playwright import urllib.request js_urls = [] def on_response(resp): url = resp.url # Capture JS bundle URLs if url.endswith(".js") and "stake.com" in url and "_nuxt" in url: js_urls.append(url) # Capture any successful API responses if "_api/graphql" in url and resp.status == 200: try: body = resp.json() op = None rb = resp.request.post_data if rb: try: rb_parsed = json.loads(rb) op = rb_parsed.get("operationName") if isinstance(rb_parsed, dict) else None except: pass captured["api_responses"].append({ "operation": op, "status": resp.status, "response": body }) print(f" [API 200!] op={op}") except: pass with sync_playwright() as p: print("[*] Firefox launch") browser = p.firefox.launch(headless=False) ctx = browser.new_context( viewport={"width": 1366, "height": 768}, user_agent="Mozilla/5.0 (X11; Linux x86_64; rv:122.0) Gecko/20100101 Firefox/122.0", ) page = ctx.new_page() page.on("response", on_response) print("[*] Loading homepage ...") try: page.goto("https://stake.com/", timeout=40000, wait_until="networkidle") except Exception as e: print(f" warn: {e}") # Wait for CF to clear for i in range(10): title = page.title().lower() if "just a moment" not in title: break print(f" CF wait {i*4}s ...") time.sleep(4) print(f" Page: {page.url} | {page.title()}") time.sleep(5) # Grab cookies BEFORE navigating further all_cookies = ctx.cookies() stake_cookies = {c["name"]: c["value"] for c in all_cookies if "stake" in c.get("domain", "")} captured["cookies"] = stake_cookies print(f" Got cookies: {list(stake_cookies.keys())}") # Grab the page's HTML to find __NUXT__ state or config print("\n[*] Checking page state ...") try: state = page.evaluate(""" () => { // Try various global state objects const candidates = [ window.__NUXT__, window.__INITIAL_STATE__, window.__APP_STATE__, ]; for (const c of candidates) { if (c) { try { return JSON.stringify(c, null, 2).slice(0, 10000); } catch(e) { return String(c).slice(0, 1000); } } } // Try to find config object const scripts = Array.from(document.querySelectorAll('script')).map(s => s.textContent); for (const s of scripts) { if (s.includes('graphql') || s.includes('_api')) { return 'FOUND_IN_SCRIPT: ' + s.slice(0, 2000); } } return null; } """) if state: print(f" State found: {state[:500]}") captured["page_state"] = state except Exception as e: print(f" warn state: {e}") # Get all script src URLs from the page print("\n[*] Collecting JS bundle URLs ...") try: script_urls = page.evaluate(""" () => Array.from(document.querySelectorAll('script[src]')) .map(s => s.src) .filter(s => s.includes('stake.com')) """) for url in script_urls: if url not in js_urls: js_urls.append(url) print(f" Found {len(js_urls)} JS bundles from script tags") except Exception as e: print(f" warn scripts: {e}") # Also try navigating to sports page now that we have cf_clearance print("\n[*] Trying sports page with cf_clearance cookie ...") try: page.goto("https://stake.com/sports/soccer", timeout=40000, wait_until="domcontentloaded") time.sleep(3) for _ in range(8): title = page.title().lower() if "just a moment" not in title: print(f" Sports page loaded: {page.url} | {page.title()}") break print(f" still CF ...") time.sleep(5) except Exception as e: print(f" warn sports: {e}") time.sleep(10) # give network time final_cookies = ctx.cookies() captured["cookies"] = {c["name"]: c["value"] for c in final_cookies if "stake" in c.get("domain", "")} print(f" Final cookies: {list(captured['cookies'].keys())}") browser.close() # ── Now download and parse JS bundles for GraphQL queries ───────────────── print(f"\n[*] Analyzing {len(js_urls)} JS bundles for GraphQL queries ...") # Build cookie header cookie_header = "; ".join(f"{k}={v}" for k, v in captured["cookies"].items()) gql_pattern = re.compile( r'(?:query|mutation|subscription)\s+(\w+)[^`"\']*[`"\']?\s*\{[^}]+\}', re.DOTALL ) op_name_pattern = re.compile(r'operationName["\s:]+["\'](\w+)["\']') query_string_pattern = re.compile(r'(?:query|mutation)\s+(\w+)\s*[\(\{]') all_op_names = set() for i, url in enumerate(js_urls[:30]): # limit to first 30 bundles try: req = urllib.request.Request(url, headers={ "User-Agent": "Mozilla/5.0 (X11; Linux x86_64; rv:122.0) Gecko/20100101 Firefox/122.0", "Cookie": cookie_header, "Referer": "https://stake.com/", }) with urllib.request.urlopen(req, timeout=15) as resp: content = resp.read().decode("utf-8", errors="replace") # Find operation names ops = op_name_pattern.findall(content) query_names = query_string_pattern.findall(content) if ops or query_names: bundle_ops = set(ops) | set(query_names) all_op_names |= bundle_ops print(f" Bundle {i+1}: {url.split('/')[-1][:30]:30} -> {len(bundle_ops)} ops: {list(bundle_ops)[:8]}") # Extract full query strings around operation names for op in ops[:5]: # Find surrounding context idx = content.find(op) if idx > 0: start = max(0, idx - 200) end = min(len(content), idx + 500) context_str = content[start:end] captured["gql_queries_found"].append({ "bundle": url.split("/")[-1], "operation": op, "context": context_str, }) # Also look for the API URL if "_api/graphql" in content: print(f" ** Contains graphql endpoint reference **") idx = content.find("_api/graphql") print(f" context: {content[max(0,idx-100):idx+100]}") except Exception as e: pass # many bundles will 404 or fail print(f"\n[*] Total unique operation names found: {len(all_op_names)}") for op in sorted(all_op_names): print(f" - {op}") captured["gql_operation_names"] = list(all_op_names) out = "/home/cyborg/Desktop/claude/arb_bot/tools/stake_discovery_results.json" with open(out, "w") as f: json.dump(captured, f, indent=2, default=str) print(f"\n[*] Saved to {out}") return captured if __name__ == "__main__": data = run() print("\n" + "="*70) print("OPERATION NAMES FOUND IN JS BUNDLES (sports-related)") print("="*70) sports_kw = ["sport", "soccer", "football", "basket", "tennis", "match", "event", "fixture", "bet", "market", "odd", "league", "live", "prematch", "upcoming", "odds"] for op in sorted(data.get("gql_operation_names", [])): if any(kw in op.lower() for kw in sports_kw): print(f" ** {op}") else: print(f" {op}") print("\n" + "="*70) print("QUERY CONTEXTS (sports-related)") print("="*70) for q in data.get("gql_queries_found", []): op = q.get("operation", "") if any(kw in op.lower() for kw in sports_kw): print(f"\n [{op}] from {q['bundle']}") print(f" {q['context'][:400]}")