🤝 Slack API Setup

The Slack API lets your bot read/post messages and automate tasks. The Events API sends payloads when events (e.g., messages) happen. 🎉

1. Create a Slack app at api.slack.com. 🛠️
2. Add these bot token scopes:
   • app_mentions:read 📣: Detects @slackAgent mentions.
   • channels:history 📜: Reads public channel message history.
   • channels:join 🚪: Joins public channels.
   • channels:read 👀: Accesses public channel details.
   • chat:write ✍️: Sends messages in channels/DMs.
   • chat:write.public 📢: Posts in public channels without joining.
   • commands ⚡: Supports custom slash commands.
   • im:history 💬: Reads DM history.
   • im:write 📩: Sends DMs.
   • mpim:history 👥: Reads group DM history.
3. Enable Event Subscriptions with your ngrok URL (e.g., https://<random>.ngrok.io/slack/events). 🔗
4. Subscribe to bot events:
   • message.channels 📢: Triggers on public channel messages.
   • message.im 💌: Triggers on direct messages.
5. Verify the setup:
   • In n8n, set a webhook node with a challenge field for Slack’s verification.
   • Set webhook to "Respond to WebHook" mode, then switch to "Immediately" after verification.
6. Add the bot (slackAgentApp) to your Slack workspace. 🤖

⚙️ n8n Workflow Setup

Automate with n8n! 🛠️ Visualize the workflow:

1. Create an n8n workflow with these nodes:
   • Webhook 🌐: Captures Slack events and verifies challenges.
   • Set Node 📝: Extracts user_query and channel_id.
   • Cleanup Node 🧹: Removes mentions (e.g., @slackAgentApp).
   • HTTP Server Node 🔗: Sends queries to FastAPI via ngrok.
   • Merge Node 🔄: Combines query and response into JSON.
   • Slack Node 💬: Sends responses to your Slack workspace.
2. Authenticate the Slack node with your workspace. 🔑
3. Run the workflow and test with a Slack query (e.g., @slackAgentApp how to install dependencies). 🚀

🌐 Streamlit UI Setup (Alternative)

Skip Slack and use a web interface! 🌍 Check out the UI:

1. Create an n8n workflow with:
   • Webhook 📡: Passes queries from Streamlit to FastAPI.
   • HTTP Server Node 🔗: Uses ngrok to expose the server.
   • Respond to Webhook 📤: Formats responses for Streamlit.
2. Run the Streamlit app:

   streamlit run chatbot_frontend.py
   

🎥 Demo

See slackAgent in action! 🚀

🎮 Usage

Chat with the bot in Slack or use the Streamlit UI! 💬

• Slack: Ping the bot (e.g., @slackAgentApp how to install dependencies).
• Streamlit: Open the web UI, type a query, and get answers! 🌐

🧪 Example

Slack Query:

@slackAgentApp How do I install dependencies?

Response:

Run `pip install -r requirements.txt` in your virtual environment. 📦

🧠 Technical Details

• Document Indexing (app.py):
  • Loads files from documentation/ with SimpleDirectoryReader. 📄
  • Creates OpenAI embeddings via VectorStoreIndex. 🧠
  • Stores in ChromaDB at storage/developer_documents_collection. 💾
• Query Handling (main.py):
  • FastAPI /query endpoint processes questions. 🚀
  • Uses precomputed index from storage/ for fast responses. ⚡

🤝 Contributing

Love slackAgent? Contribute! 🙌 Open an issue or submit a pull request to make it even better. 🌟