Documentation Index
Fetch the complete documentation index at: https://docs.enconvo.ai/llms.txt
Use this file to discover all available pages before exploring further.
Overview
The IM Channels extension connects EnConvo to your team’s messaging platforms. Send and read messages on Slack, Discord, Telegram, and Feishu (Lark) directly through AI conversations, or set up automated bot listeners that reply to incoming messages using your configured LLM. This turns EnConvo into a unified messaging hub where your AI assistant can communicate across platforms on your behalf.Slack
Send messages, read channels, Socket Mode listener
Discord
Channel messaging via REST API and Gateway WebSocket
Telegram
Bot API messaging with long-polling listener
Feishu / Lark
Open API integration with auto-refreshing tokens
Setting Up Channels
Each platform requires its own bot token or API credentials. Configure them in the extension settings.Slack
Create a Slack App
Go to api.slack.com/apps and create a new app for your workspace
Configure Bot Permissions
Under OAuth & Permissions, add the scopes your bot needs (e.g.,
chat:write, channels:read, channels:history)Install to Workspace
Install the app to your Slack workspace and copy the Bot User OAuth Token (
xoxb-...)Discord
Create a Discord Application
Go to the Discord Developer Portal and create a new application
Set Permissions
Under OAuth2 -> URL Generator, select the
bot scope and the permissions you need (Send Messages, Read Message History, etc.)Telegram
Create a Bot via BotFather
Open Telegram and message @BotFather. Send
/newbot and follow the prompts to create your botFeishu / Lark
Create a Feishu App
Go to the Feishu Open Platform and create a new application
Configure Permissions
Add the messaging permissions your bot needs (send messages, read messages, etc.)
Enter Credentials in EnConvo
Go to Settings -> IM Channels -> Feishu / Lark Channel and enter the App ID and App Secret
Using Messaging in Conversations
Once channels are configured, the AI can send and read messages on your behalf.Sending Messages
Reading Messages
Cross-Platform Communication
Bot Listener
The Channel Listener Service runs in the background and automatically replies to incoming messages using your configured AI model.How It Works
- The listener connects to each configured platform (Slack via Socket Mode, Discord via Gateway WebSocket, Telegram via long-polling)
- When someone messages or mentions your bot, the listener receives the message
- The message is processed by your configured LLM provider
- The AI’s response is sent back to the channel
Agent Binding
You can bind specific channels to specific AI agents. This lets different channels use different system prompts, tools, and behaviors.| Concept | Description |
|---|---|
| One agent to many channels | A single agent can serve multiple channels across platforms |
| One channel to one agent | Each channel is bound to exactly one agent |
| Default fallback | Unbound channels use the default LLM configuration |
Configuring the Listener
- Go to Settings -> IM Channels -> Channel Listener Service
- Select the AI Model Provider to use for bot replies
- The listener automatically starts for all configured channels with valid credentials
Platform Details
| Platform | Transport | Listener Method | Max Message Length |
|---|---|---|---|
| Slack | HTTPS Web API | Socket Mode WebSocket | 4,000 characters |
| Discord | HTTPS REST v10 | Gateway WebSocket | 2,000 characters |
| Telegram | HTTPS Bot API | Long-polling (30s) | 4,096 characters |
| Feishu | HTTPS Open API | Polling | 4,000 characters |
Long messages are automatically split into multiple parts at natural break points (newlines, then spaces) to stay within platform limits.
Using Channels in Workflows
Messaging channels integrate with EnConvo workflows for automated communication:| Workflow | Description |
|---|---|
| Status Broadcast | Automatically post status updates to multiple channels |
| Alert Pipeline | Monitor events and send alerts to the appropriate channel |
| Daily Digest | Compile information and post a daily summary |
| Cross-Platform Relay | Read from one platform and forward to another |
Troubleshooting
Channel shows as not configured
Channel shows as not configured
Verify that the bot token or API credentials are entered correctly in Settings -> IM Channels. Tokens must be valid and not expired.
Bot cannot send messages
Bot cannot send messages
Ensure the bot has been invited to the target channel/group and has the correct permissions. On Slack, the bot must be added to channels before it can post.
Listener not responding to messages
Listener not responding to messages
Check that the Channel Listener Service is running. For Slack, Socket Mode requires an App-Level Token. For Discord, the bot needs the Message Content intent enabled in the Developer Portal.
Messages are cut off
Messages are cut off
Each platform has a maximum message length. EnConvo automatically splits long messages, but very long responses may be sent across multiple messages.
Feishu token errors
Feishu token errors
Feishu tokens auto-refresh, but if you see authentication errors, verify your App ID and App Secret. For Lark international, make sure the API Domain is set to
https://open.larksuite.com.Related Features
Gmail Integration
Email management via Gmail API
Apple Ecosystem
Native Apple Mail, Reminders, Calendar
Workflows
Automate messaging in workflows
Agents
Bind AI agents to messaging channels