import os # Telegram Bot BOT_TOKEN = os.environ.get("BOT_TOKEN", "") # AI provider — OpenAI-compatible (SiliconFlow / OpenAI / DeepSeek / local vLLM / any relay) AI_API_KEY = os.environ.get("AI_API_KEY", "") AI_BASE_URL = os.environ.get("AI_BASE_URL", "https://api.siliconflow.cn/v1") AI_MODEL = os.environ.get("AI_MODEL", "deepseek-ai/DeepSeek-V4-Flash") AI_VISION_MODELS = os.environ.get( "AI_VISION_MODELS", "Qwen/Qwen3-VL-30B-A3B-Instruct,Qwen/Qwen3-VL-32B-Instruct,Qwen/Qwen3-VL-8B-Instruct", ) AI_AUDIO_MODEL = os.environ.get("AI_AUDIO_MODEL", "Qwen/Qwen3-Omni-30B-A3B-Instruct") AI_IMAGE_MAX_WIDTH = int(os.environ.get("AI_IMAGE_MAX_WIDTH", "600")) # Optional second AI provider, tried only after every primary model failed. # Leave AI_FALLBACK_API_KEY empty and nothing changes: the chain stays exactly # as it was. Worth setting if you have a spare relay/vendor key -- when the # primary times out the judge otherwise falls back to the keyword lexicon # alone, which only knows the slang already in the wordlist. AI_FALLBACK_API_KEY = os.environ.get("AI_FALLBACK_API_KEY", "") AI_FALLBACK_BASE_URL = os.environ.get("AI_FALLBACK_BASE_URL", "") AI_FALLBACK_MODEL = os.environ.get("AI_FALLBACK_MODEL", "") AI_FALLBACK_VISION_MODEL = os.environ.get("AI_FALLBACK_VISION_MODEL", "") # Database DB_PATH = os.environ.get("DB_PATH", "/data/bot.db") # Anti-spam SPAM_NEWUSER_COOLDOWN = int(os.environ.get("SPAM_NEWUSER_COOLDOWN", "300")) SPAM_REPEAT_THRESHOLD = int(os.environ.get("SPAM_REPEAT_THRESHOLD", "3")) SPAM_REPEAT_WINDOW = int(os.environ.get("SPAM_REPEAT_WINDOW", "300")) # Chinese-slang lexicon hard-hit threshold (prefilter Layer 2.5). Higher = stricter. LEXICON_HARD_THRESHOLD = int(os.environ.get("LEXICON_HARD_THRESHOLD", "6")) # Probe ("check-in") detection. Mark-only layer: it never bans, it only starts # silently deleting once a member has been flagged this many times... PROBE_FLAG_DELETE = int(os.environ.get("PROBE_FLAG_DELETE", "2")) # ...within this rolling window (hours). Flags older than the window reset to 1, # so a real person posting one such message a day is never permanently silenced. PROBE_WINDOW_HOURS = int(os.environ.get("PROBE_WINDOW_HOURS", "24")) # How much of an over-long message the spam judge sees. The text is clipped # head + tail (never head only), so an ad appended to the end of a long paste # still lands inside the judged range. JUDGE_TEXT_LIMIT = int(os.environ.get("JUDGE_TEXT_LIMIT", "1800")) JUDGE_CAPTION_LIMIT = int(os.environ.get("JUDGE_CAPTION_LIMIT", "900")) # Public group commands reply with reply_text, which bumps the triggering # message to the top of the chat. Without a cooldown anyone can spam a command # and have the bot repeatedly bump their own ad. Seconds, per chat, non-admins # only — admins and private chats are never limited. GROUP_CMD_COOLDOWN = int(os.environ.get("GROUP_CMD_COOLDOWN", "60")) # Super-admin Telegram user ID, for /groups. Must be set explicitly: this used to # be claimed by whoever sent the first /start after a restart, which on a public # bot means a stranger, re-opened on every deploy. Unset = nobody has it. try: ADMIN_USER_ID = int(os.environ.get("ADMIN_USER_ID", "") or 0) or None except ValueError: ADMIN_USER_ID = None