````md
# Slack → Dify Chatbot (Socket Mode)

This project is a small Slack bot that listens for **@mentions** and forwards the user’s message to a **Dify agent**.
It then posts Dify’s response back into the **same Slack thread**.

It also keeps **conversation context per Slack thread**, so replies in the same thread continue the same Dify conversation.

---

## What it does

- ✅ Listens for `app_mention` events in Slack (when someone types `@yourbot ...`)
- ✅ Removes the Slack mention token (`<@U123...>`) from the text
- ✅ Sends the cleaned text to Dify’s `POST /v1/chat-messages` endpoint
- ✅ Uses **streaming mode** (SSE) and parses the `data: {...}` lines into JSON events
- ✅ Chooses the **latest** `agent_thought` event and posts its `thought` to Slack
  - If no `agent_thought` is found, it falls back to the streamed `agent_message` chunks
- ✅ Stores a mapping of:
  - `Slack thread (channel + thread_ts)` → `Dify conversation_id`

---

## Requirements

- Python 3.10+ recommended
- A Slack App configured for **Socket Mode**
- A Dify instance (cloud or self-hosted) with an API key

---

## Setup locally

### 1) Clone and enter the folder
```bash
git clone <your-repo-url>
cd slack_chatbot
````

### 2) Create and activate a virtual environment (Windows example)

```powershell
python -m venv .venv
.\.venv\Scripts\activate
```

### 3) Install dependencies

```bash
pip install -r requirements.txt
```

### 4) Create your `.env` file

In the same folder as `main.py`, create a file named `.env`:

```env
APP_TOKEN=xapp-...
BOT_TOKEN=xoxb-...
DIFY_API_KEY=...
DIFY_BASE_URL=https://api.dify.ai
```

> The code loads `.env` from the same directory as `main.py`.

### 5) Run the bot

```bash
python main.py
```

If everything is correct, the bot will connect to Slack via Socket Mode and start listening.

---

## Slack App configuration (quick checklist)

In your Slack App settings:

1. **Socket Mode**

   * Enable Socket Mode
   * Generate an **App-Level Token** with the scope: `connections:write`
   * This token is your `APP_TOKEN` (starts with `xapp-...`)

2. **OAuth & Permissions**

   * Add these bot token scopes (common minimum):

     * `app_mentions:read`
     * `chat:write`
   * Install the app to your workspace
   * Copy the **Bot User OAuth Token** as `BOT_TOKEN` (starts with `xoxb-...`)

3. **Event Subscriptions**

   * Enable Events
   * Subscribe to **Bot Events**:

     * `app_mention`

---

## Environment variables

These variables must exist in your `.env`:

| Variable        | Example / Notes                                                          |
| --------------- | ------------------------------------------------------------------------ |
| `APP_TOKEN`     | Slack Socket Mode app token (`xapp-...`)                                 |
| `BOT_TOKEN`     | Slack bot token (`xoxb-...`)                                             |
| `DIFY_API_KEY`  | Dify API key (Bearer token)                                              |
| `DIFY_BASE_URL` | Base URL for Dify, e.g. `https://api.dify.ai` or your self-hosted domain |

---

## How replies and context work

Slack threads are used like “chat sessions”.

* Each Slack thread has a unique `thread_ts`
* The bot stores a mapping:

  * `channel:thread_ts` → `conversation_id`
* So if you keep talking in the same Slack thread, the bot continues the same Dify conversation.

> Note: this mapping is currently stored **in memory**, so restarting the bot resets it.

---

## Security notes

* ✅ `.env` should NOT be committed to GitHub
* Add `.env` to your `.gitignore` (recommended)

You can optionally commit a safe template file named `.env.example` so others know which variables are required.

---

## Troubleshooting

### `KeyError: 'APP_TOKEN'` (or other env var)

* Make sure you have a `.env` file next to `main.py`
* Make sure you installed `python-dotenv`
* Make sure the variable names match exactly (`APP_TOKEN`, not `APP-TOKEN`, etc.)

### Bot connects but doesn’t respond

* Confirm your Slack App has the `app_mention` event subscribed
* Confirm the bot has `chat:write` scope and is installed to the workspace
* Mention the bot like: `@yourbot hello`

---

## License

Use however you like (add your preferred license here).

```
```