hack/tools/scripts/llm-key-hunter/verify.py

#!/usr/bin/env python3
"""Fast parallel LLM API key verifier — Python version.

Uses urllib instead of spawning curl subprocesses per key.
"""

import time
from pathlib import Path
from concurrent.futures import ThreadPoolExecutor, as_completed
from collections import defaultdict
import urllib.request
import urllib.error

RESULTS_DIR = Path(__file__).parent / "results"
KEYS_FILE = RESULTS_DIR / "extracted_keys.txt"
LIVE_FILE = RESULTS_DIR / "live_keys_raw.txt"

# Provider verify configs
VERIFY_CONFIGS = {
    "OpenAI":       {"url": "https://api.openai.com/v1/models", "headers": {"Authorization": "Bearer {key}"}},
    "Anthropic":    {"url": "https://api.anthropic.com/v1/models", "headers": {"x-api-key": "{key}", "anthropic-version": "2023-06-01"}},
    "Google":       {"url": "https://generativelanguage.googleapis.com/v1/models?key={key}", "headers": {}},
    "HuggingFace":  {"url": "https://huggingface.co/api/whoami-v2", "headers": {"Authorization": "Bearer {key}"}},
    "Groq":         {"url": "https://api.groq.com/openai/v1/models", "headers": {"Authorization": "Bearer {key}"}},
    "Replicate":    {"url": "https://api.replicate.com/v1/account", "headers": {"Authorization": "Token {key}"}},
    "TogetherAI":   {"url": "https://api.together.xyz/models", "headers": {"Authorization": "Bearer {key}"}},
    "DeepSeek":     {"url": "https://api.deepseek.com/v1/models", "headers": {"Authorization": "Bearer {key}"}},
    "OpenRouter":   {"url": "https://openrouter.ai/api/v1/models", "headers": {"Authorization": "Bearer {key}"}},
    "Perplexity":   {"url": "https://api.perplexity.ai/chat/completions", "headers": {"Authorization": "Bearer {key}"}},
    "VolcanoArk":   {"url": "https://ark.cn-beijing.volces.com/api/v3/models", "headers": {"Authorization": "Bearer {key}"}},
    "ZhipuAI":      {"url": "https://open.bigmodel.cn/api/paas/v4/models", "headers": {"Authorization": "Bearer {key}"}},
    "DashScope":    {"url": "https://dashscope.aliyuncs.com/compatible-mode/v1/models", "headers": {"Authorization": "Bearer {key}"}},
    "Moonshot":     {"url": "https://api.moonshot.cn/v1/models", "headers": {"Authorization": "Bearer {key}"}},
    "MiniMax":      {"url": "https://api.minimaxi.com/v1/models", "headers": {"Authorization": "Bearer {key}"}},
    "LingyiWanwu":  {"url": "https://api.lingyiwanwu.com/v1/models", "headers": {"Authorization": "Bearer {key}"}},
    "StepFun":      {"url": "https://api.stepfun.com/v1/models", "headers": {"Authorization": "Bearer {key}"}},
    "SiliconFlow":  {"url": "https://api.siliconflow.cn/v1/models", "headers": {"Authorization": "Bearer {key}"}},
    "LongCat":      {"url": "https://api.longcat.chat/openai/models", "headers": {"Authorization": "Bearer {key}"}},
    "OllamaCloud":  {"url": "https://api.ollama.com/api/tags", "headers": {"Authorization": "Bearer {key}"}},
    "FreeModel":    {"url": "https://api.freemodel.dev/v1/models", "headers": {"Authorization": "Bearer {key}"}},
    "XKiro":        {"url": "https://api.xkiro.com/v1/models", "headers": {"Authorization": "Bearer {key}"}},
}

# Providers to skip (too many false positives, or no simple verify endpoint)
SKIP_PROVIDERS = {
    "Mistral", "Cohere", "AzureOpenAI",
    "iFlytekSpark", "iFlytekCodingPlan",
    "TencentHunyuan",
    "BaiduQianfan",
    "CSDNCodingPlan", "HuaweiCodeArts", "MiMo", "InfiniAI", "JDCloud",
    "MooreThreads", "Kuaishou", "UCloud", "Anomaly", "ChinaUnicom",
    "SCNet", "AlibabaCodingPlan", "Zyloo",
}


def verify_key(provider, key):
    """Verify a key against its provider API. Returns True if live, False if dead, None if skipped."""
    if provider in SKIP_PROVIDERS:
        return None
    if provider not in VERIFY_CONFIGS:
        return None

    cfg = VERIFY_CONFIGS[provider]
    url = cfg["url"].replace("{key}", key)
    headers = {hname: hval.replace("{key}", key) for hname, hval in cfg["headers"].items()}

    req = urllib.request.Request(url, headers=headers, method="GET")
    try:
        with urllib.request.urlopen(req, timeout=10) as resp:
            code = resp.getcode()
            return code == 200 or (provider == "Perplexity" and code not in (401, 0))
    except urllib.error.HTTPError as e:
        code = e.code
        if provider == "Perplexity":
            return code not in (401, 0)
        return code == 200
    except Exception:
        return False


def main():
    # Read keys
    keys = []
    with open(KEYS_FILE) as f:
        for line in f:
            line = line.strip()
            if not line:
                continue
            parts = line.split("|", 3)
            if len(parts) >= 2:
                provider = parts[0]
                key = parts[1]
                url = parts[2] if len(parts) > 2 else ""
                desc = parts[3] if len(parts) > 3 else ""
                keys.append((provider, key, url, desc))

    print(f"Total keys: {len(keys)}")

    # Group by provider and deduplicate
    by_provider = defaultdict(dict)  # provider -> {key: (url, desc)}
    for provider, key, url, desc in keys:
        if provider in SKIP_PROVIDERS:
            continue
        if provider not in VERIFY_CONFIGS:
            continue
        if key not in by_provider[provider]:
            by_provider[provider][key] = (url, desc)

    # Print summary
    print("\nKeys by provider (deduplicated):")
    total_to_verify = 0
    for provider in sorted(by_provider.keys()):
        n = len(by_provider[provider])
        total_to_verify += n
        print(f"  {provider:20s}: {n:6d}")
    print(f"  {'TOTAL':20s}: {total_to_verify:6d}")

    # Build verify list
    to_verify = []
    for provider, keydict in by_provider.items():
        for key, (url, desc) in keydict.items():
            to_verify.append((provider, key, url, desc))

    # Verify in parallel
    live_keys = []
    dead_count = 0
    processed = 0
    start_time = time.time()

    print(f"\nVerifying {len(to_verify)} keys with 30 threads...")

    with ThreadPoolExecutor(max_workers=30) as pool:
        futures = {pool.submit(verify_key, p, k): (p, k, u, d) for p, k, u, d in to_verify}

        for future in as_completed(futures):
            provider, key, url, desc = futures[future]
            processed += 1

            try:
                result = future.result()
            except Exception:
                result = False

            if result is True:
                live_keys.append((provider, key, url, desc))
            elif result is False:
                dead_count += 1

            if processed % 1000 == 0:
                elapsed = time.time() - start_time
                rate = processed / elapsed
                print(f"  [{processed}/{len(to_verify)}] live={len(live_keys)} dead={dead_count} ({rate:.0f}/s)")

    elapsed = time.time() - start_time
    print(f"\nDone in {elapsed:.1f}s: {len(live_keys)} live, {dead_count} dead")

    # Write results
    with open(LIVE_FILE, "w") as f:
        for provider, key, url, desc in sorted(live_keys):
            f.write(f"LIVE|{provider}|{key}|{url}|{desc}\n")

    print(f"Live keys written to: {LIVE_FILE}")

    # Print live keys by provider
    live_by_provider = defaultdict(int)
    for provider, key, url, desc in live_keys:
        live_by_provider[provider] += 1

    print("\nLive keys by provider:")
    for provider in sorted(live_by_provider.keys()):
        print(f"  {provider:20s}: {live_by_provider[provider]}")


if __name__ == "__main__":
    main()