project-root/ setup.py README.md app/ init.py config.py llm.py main.py
from setuptools import setup, find_packages
setup( name="slack_hf_bot", version="0.1.0", description="Slack chatbot using FastAPI + Hugging Face Inference Providers", author="Your Name", packages=find_packages(exclude=("tests",)), install_requires=[ "fastapi", "uvicorn[standard]", "slack-sdk", "httpx", ], python_requires=">=3.10", )
A production-ready Slack chatbot powered by FastAPI and Hugging Face’s OpenAI-compatible /v1/chat/completions endpoint.
This bot:
- Listens to Slack
app_mentionevents - Verifies Slack signatures
- Sends your message to an LLM (Gemma/Phi/etc.)
- Replies inside Slack with AI answers
python -m venv venv source venv/bin/activate # macOS / Linux venv\Scripts\activate # Windows
pip install --upgrade pip pip install -e .
Go to https://api.slack.com/apps → Create New App.
Slack → Basic Information
- Signing Secret → put in
Config.slack_signing_secret
Slack → OAuth & Permissions Add Bot Token Scopes:
- app_mentions:read
- chat:write
Install to workspace → copy Bot User OAuth Token (xoxb-...)
Put in Config.slack_bot_token
Slack → Event Subscriptions
- Enable events ON
- Add bot event:
app_mention - Request URL will be added after ngrok is running
- Go to https://huggingface.co/settings/tokens
- Create token with permission:
Make calls to Inference Providers - Copy token → put in
Config.hf_token
Choose a free model:
- google/gemma-2-2b-it
- microsoft/Phi-3.5-mini-instruct
Put in: Config.hf_model
uvicorn app.main:app --reload
Check health: http://127.0.0.1:8000/health
ngrok config add-authtoken <YOUR_TOKEN> ngrok http 8000
Copy the HTTPS URL:
Set in Slack → Event Subscriptions → Request URL:
https://xxxxxx.ngrok-free.app/slack/events
→ should show “Verified”
In any channel:
@YourBotName hello
You should get an AI response.
class Config: # Slack credentials (replace these) slack_signing_secret = "REPLACE_WITH_SLACK_SIGNING_SECRET" slack_bot_token = "xoxb-REPLACE_WITH_SLACK_BOT_TOKEN"
# Hugging Face Inference Providers
hf_token = "hf_REPLACE_WITH_HF_TOKEN"
# Free LLM models such as:
# google/gemma-2-2b-it
# microsoft/Phi-3.5-mini-instruct
hf_model = "google/gemma-2-2b-it"
import httpx from .config import Config
HF_TOKEN = Config.hf_token HF_MODEL = Config.hf_model HF_URL = "https://router.huggingface.co/v1/chat/completions"
async def llm_echo(text: str) -> str: return f"Model: {HF_MODEL}\nECHO: {text}"
async def test_llm_connection(): try: headers = { "Authorization": f"Bearer {HF_TOKEN}", "Content-Type": "application/json", } payload = { "model": HF_MODEL, "messages": [{"role": "user", "content": "ping"}], "stream": False, }
async with httpx.AsyncClient(timeout=15) as client:
r = await client.post(HF_URL, headers=headers, json=payload)
r.raise_for_status()
return "HF_ROUTER_OK"
except Exception as e:
return f"ERROR: {e}"
async def llm_generate(text: str) -> str: headers = { "Authorization": f"Bearer {HF_TOKEN}", "Content-Type": "application/json", } payload = { "model": HF_MODEL, "messages": [{"role": "user", "content": text}], "stream": False, "max_tokens": 512, "temperature": 0.7, "top_p": 0.9, }
async with httpx.AsyncClient(timeout=30) as client:
r = await client.post(HF_URL, headers=headers, json=payload)
r.raise_for_status()
data = r.json()
return data["choices"][0]["message"]["content"]
import json import time import hmac import hashlib from fastapi import FastAPI, Request, HTTPException, BackgroundTasks from slack_sdk import WebClient from slack_sdk.errors import SlackApiError
from .config import Config from .llm import llm_echo, test_llm_connection, llm_generate
app = FastAPI(title="SlackBot HF", version="1.1")
SLACK_SIGNING_SECRET = Config.slack_signing_secret SLACK_BOT_TOKEN = Config.slack_bot_token
client = WebClient(token=SLACK_BOT_TOKEN)
def verify_slack(req: Request, raw_body: bytes): timestamp = req.headers.get("X-Slack-Request-Timestamp") signature = req.headers.get("X-Slack-Signature")
if not timestamp or not signature:
raise HTTPException(status_code=401, detail="Missing Slack headers")
if abs(time.time() - int(timestamp)) > 300:
raise HTTPException(status_code=401, detail="Stale timestamp")
basestring = f"v0:{timestamp}:{raw_body.decode()}"
computed = "v0=" + hmac.new(
SLACK_SIGNING_SECRET.encode(),
basestring.encode(),
hashlib.sha256,
).hexdigest()
if not hmac.compare_digest(computed, signature):
raise HTTPException(status_code=401, detail="Invalid signature")
@app.get("/health") def health(): return {"ok": True}
@app.post("/slack/events") async def slack_events(req: Request, bg: BackgroundTasks): raw_body = await req.body()
try:
verify_slack(req, raw_body)
except Exception as e:
raise HTTPException(status_code=401, detail=str(e))
data = json.loads(raw_body.decode() or "{}")
if data.get("type") == "url_verification":
return {"challenge": data["challenge"]}
if data.get("type") == "event_callback":
event = data["event"]
if "bot_id" in event:
return {"ok": True}
user = event.get("user")
channel = event.get("channel")
text = event.get("text", "")
if "<@" in text:
text = text.split(">", 1)[1].strip()
async def reply():
echo_msg = await llm_echo(text)
answer = await llm_generate(text)
llm_status = await test_llm_connection()
msg = (
f"👋 Hello <@{user}>!\n"
f"*Echo:* {echo_msg}\n\n"
f"*Answer:*\n{answer}\n\n"
f"LLM Status: `{llm_status}`"
)
client.chat_postMessage(channel=channel, text=msg)
bg.add_task(reply)
return {"ok": True}