ilang-ai
fix: pass BLOCK_NONE safety_settings on every API call, not just constructor
4abec07
import json
import logging
import random
import os
import google.generativeai as genai
import config
logger = logging.getLogger(__name__)
genai.configure(api_key=config.GEMINI_API_KEY)
# Relax Gemini safety filters
SAFETY_SETTINGS = [
{"category": "HARM_CATEGORY_HARASSMENT", "threshold": "BLOCK_NONE"},
{"category": "HARM_CATEGORY_HATE_SPEECH", "threshold": "BLOCK_NONE"},
{"category": "HARM_CATEGORY_SEXUALLY_EXPLICIT", "threshold": "BLOCK_NONE"},
{"category": "HARM_CATEGORY_DANGEROUS_CONTENT", "threshold": "BLOCK_NONE"},
]
# Load prompts from .ilang files (prompts/ if exists, else prompts_demo/)
def _load_prompt(name):
for d in ("prompts", "prompts_demo"):
path = os.path.join(os.path.dirname(os.path.dirname(__file__)), d, name)
if os.path.exists(path):
with open(path, "r", encoding="utf-8") as f:
content = f.read()
logger.info("Loaded prompt: " + path + " (" + str(len(content)) + " chars)")
return content
logger.warning("Prompt not found: " + name)
return ""
SYSTEM_PROMPT = _load_prompt("persona.ilang")
ANTISPAM_TEXT_PROMPT = _load_prompt("antispam.ilang")
VISION_PROMPT = _load_prompt("vision.ilang")
if not SYSTEM_PROMPT:
SYSTEM_PROMPT = "You are TelegramGuard, a helpful AI assistant on Telegram. Reply concisely in the user's language. JSON format: {\"intent\": \"chat\", \"device\": null, \"reply\": \"your text\"}"
logger.warning("Using fallback system prompt")
GROUP_WELCOME = (
"I-Lang Guard is here\n\n"
"I auto-clean spam. No config needed.\n"
"@ me if you need anything.\n\n"
"Just give me admin permissions (delete messages + ban users)."
)
model = genai.GenerativeModel(
config.GEMINI_MODEL,
system_instruction=SYSTEM_PROMPT,
safety_settings=SAFETY_SETTINGS
)
vision_model = genai.GenerativeModel(
config.GEMINI_MODEL,
safety_settings=SAFETY_SETTINGS
)
def _parse(raw):
if not raw:
return ("chat", None, "...")
t = raw.strip()
if t.startswith("```"):
nl = t.find("\n")
t = t[nl + 1:] if nl > 0 else t[3:]
if t.endswith("```"):
t = t[:-3].strip()
try:
d = json.loads(t)
return (d.get("intent", "chat"), d.get("device"), d.get("reply", t))
except json.JSONDecodeError:
pass
last_brace = t.rfind("}")
while last_brace >= 0:
start = t.rfind("{", 0, last_brace)
if start >= 0:
try:
d = json.loads(t[start:last_brace + 1])
return (d.get("intent", "chat"), d.get("device"), d.get("reply", t))
except json.JSONDecodeError:
pass
last_brace = t.rfind("}", 0, last_brace)
for line in t.split("\n"):
line = line.strip()
if line and not line.startswith("{") and not line.startswith("taint") and not line.startswith("The "):
return ("chat", None, line)
return ("chat", None, "...")
def _ctx(history, info):
parts = []
if info:
parts.append("[ctx] " + info)
if history:
for h in history[-8:]:
r = "user" if h["role"] == "user" else "bot"
parts.append(r + ": " + h["text"])
return "\n".join(parts)
def _safe_text(response):
"""Safely extract text from Gemini response — r.text throws ValueError when blocked."""
try:
if response.text:
return response.text.strip()
except (ValueError, AttributeError):
pass
# Try extracting from candidates
try:
if response.candidates:
for c in response.candidates:
if hasattr(c, 'content') and c.content and c.content.parts:
return c.content.parts[0].text.strip()
except Exception:
pass
return ""
def _deflect():
lines = [
"That's a tough one. What else can I help with?",
"Let's try a different angle. What do you need today?",
"That's beyond my range. What else is on your mind?",
]
return random.choice(lines)
async def ai_text(text, history=None, context_info=""):
try:
c = _ctx(history, context_info)
prompt = c + "\nuser: " + text if c else "user: " + text
r = await model.generate_content_async(prompt, safety_settings=SAFETY_SETTINGS)
raw = _safe_text(r)
if not raw:
feedback = ""
if hasattr(r, 'prompt_feedback'):
feedback = str(r.prompt_feedback)
if hasattr(r, 'candidates') and r.candidates:
for cand in r.candidates:
if hasattr(cand, 'finish_reason'):
feedback += " finish:" + str(cand.finish_reason)
if hasattr(cand, 'safety_ratings'):
feedback += " safety:" + str(cand.safety_ratings)
logger.warning("AI empty response. feedback=" + feedback + " prompt_len=" + str(len(prompt)))
return ("chat", None, _deflect())
logger.info("AI raw[" + str(len(raw)) + "]: " + raw[:200])
return _parse(raw)
except Exception as e:
logger.warning("AI text exception: " + str(e))
return ("chat", None, _deflect())
async def ai_vision(image_bytes, caption="", history=None, context_info=""):
try:
c = _ctx(history, context_info)
prompt = VISION_PROMPT + "\n" + c
if caption:
prompt += "\nuser: " + caption
r = await vision_model.generate_content_async([prompt, {"mime_type": "image/jpeg", "data": image_bytes}], safety_settings=SAFETY_SETTINGS)
return _parse(_safe_text(r))
except Exception as e:
logger.warning("AI vision: " + str(e))
return ("chat", None, "Couldn't read that image. Try another one?")
async def ai_voice(audio_bytes, mime_type="audio/ogg", history=None, context_info=""):
try:
c = _ctx(history, context_info)
prompt = SYSTEM_PROMPT + "\n" + c + "\nUser sent a voice message:"
r = await vision_model.generate_content_async([prompt, {"mime_type": mime_type, "data": audio_bytes}], safety_settings=SAFETY_SETTINGS)
return _parse(_safe_text(r))
except Exception as e:
logger.warning("AI voice: " + str(e))
return ("chat", None, "Didn't catch that. Try again or type it out.")
async def ai_judge_group_message(text):
try:
prompt = ANTISPAM_TEXT_PROMPT + "\n\nMessage content: " + text[:1000]
r = await vision_model.generate_content_async(prompt, safety_settings=SAFETY_SETTINGS)
result = (_safe_text(r) or "ok").lower()
return "spam" in result
except Exception:
return False
async def ai_judge_group_image(image_bytes, caption=""):
try:
prompt = ANTISPAM_TEXT_PROMPT + "\n\nJudge this image. Reply ONLY: spam or ok."
if caption:
prompt += "\nCaption: " + caption[:500]
r = await vision_model.generate_content_async([prompt, {"mime_type": "image/jpeg", "data": image_bytes}], safety_settings=SAFETY_SETTINGS)
result = (_safe_text(r) or "ok").lower()
return "spam" in result
except Exception:
return False
async def ai_group_vision(image_bytes, caption="", history=None):
try:
ctx = _ctx(history, "GROUP_CHAT: User shared an image and @mentioned you. Comment naturally in 1-2 sentences.")
prompt = SYSTEM_PROMPT + "\n" + ctx
if caption:
prompt += "\nuser: " + caption
else:
prompt += "\nuser: [shared an image]"
r = await vision_model.generate_content_async([prompt, {"mime_type": "image/jpeg", "data": image_bytes}], safety_settings=SAFETY_SETTINGS)
raw = _safe_text(r)
if not raw:
return _deflect()
intent, device, reply = _parse(raw)
if reply in ("...", ""):
return _deflect()
return reply
except Exception:
return _deflect()
async def ai_group_reply(text, history=None):
try:
ctx = _ctx(history, "GROUP_CHAT: You were @mentioned in a group. Reply directly, 1-2 sentences.")
prompt = ctx + "\nuser: " + text
r = await model.generate_content_async(prompt, safety_settings=SAFETY_SETTINGS)
raw = _safe_text(r)
if not raw:
return _deflect()
intent, device, reply = _parse(raw)
if reply in ("...", ""):
return _deflect()
return reply
except Exception:
return _deflect()