Channel Adapters

OpenFang connects to messaging platforms through 40 channel adapters, allowing users to interact with their agents across every major communication platform. Adapters span consumer messaging, enterprise collaboration, social media, community platforms, privacy-focused protocols, and generic webhooks.

All adapters share a common foundation: graceful shutdown via watch::channel, exponential backoff on connection failures, Zeroizing<String> for secrets, automatic message splitting for platform limits, per-channel model/prompt overrides, DM/group policy enforcement, per-user rate limiting, and output formatting (Markdown, TelegramHTML, SlackMrkdwn, PlainText).

All 40 Channels

Core (7)

Enterprise (8)

Social (8)

Community (6)

Self-hosted (1)

Privacy (3)

Workplace (4)

Notification (2)

Integration (1)

Channel Configuration

All channel configurations live in ~/.openfang/config.toml under the [channels] section. Each channel is a subsection:

[channels.telegram]
bot_token_env = "TELEGRAM_BOT_TOKEN"
default_agent = "assistant"
allowed_users = ["123456789"]

[channels.discord]
bot_token_env = "DISCORD_BOT_TOKEN"
default_agent = "coder"

[channels.slack]
bot_token_env = "SLACK_BOT_TOKEN"
app_token_env = "SLACK_APP_TOKEN"
default_agent = "ops"

# Enterprise example
[channels.teams]
app_id_env = "TEAMS_APP_ID"
app_secret_env = "TEAMS_APP_SECRET"
default_agent = "ops"

# Social example
[channels.mastodon]
token_env = "MASTODON_TOKEN"
instance = "https://mastodon.social"
default_agent = "social-media"

Common Fields

• bot_token_env / token_env – The environment variable holding the bot/access token. OpenFang reads the token from this env var at startup. All secrets are stored as Zeroizing<String> and wiped from memory on drop.
• default_agent – The agent name (or ID) that receives messages when no specific routing applies.
• allowed_users – Optional list of platform user IDs allowed to interact. Empty means allow all.
• overrides – Optional per-channel behavior overrides.

Environment Variables Reference (Core Channels)

Channel Overrides

Every channel adapter supports ChannelOverrides, which let you customize behavior per channel without modifying the agent manifest. Add an [channels.<name>.overrides] section in config.toml:

[channels.telegram.overrides]
model = "gemini-2.5-flash"
system_prompt = "You are a concise Telegram assistant. Keep replies under 200 words."
dm_policy = "respond"
group_policy = "mention_only"
rate_limit_per_user = 10
threading = true
output_format = "telegram_html"
usage_footer = "compact"

Override Fields

Formatter, Rate Limiter, and Policies

Output Formatter

The formatter module (openfang-channels/src/formatter.rs) converts Markdown output from the LLM into platform-native formats:

Per-User Rate Limiter

The ChannelRateLimiter (openfang-channels/src/rate_limiter.rs) uses a DashMap to track per-user message counts.

DM Policy

Controls how the adapter handles direct messages:

Group Policy

Controls how the adapter handles messages in group chats, channels, and rooms:

Telegram

Prerequisites

• A Telegram bot token (from @BotFather)

Setup

1. Open Telegram and message @BotFather.
2. Send /newbot and follow the prompts to create a new bot.
3. Copy the bot token.
4. Set the environment variable:

export TELEGRAM_BOT_TOKEN=123456789:ABCdefGHIjklMNOpqrsTUVwxyz

5. Add to config:

[channels.telegram]
bot_token_env = "TELEGRAM_BOT_TOKEN"
default_agent = "assistant"

6. Restart the daemon:

openfang start

How It Works

The Telegram adapter uses long-polling via the getUpdates API. It polls every few seconds with a 30-second long-poll timeout. On API failures, it applies exponential backoff (starting at 1 second, up to 60 seconds). Shutdown is coordinated via a watch::channel.

Discord

Prerequisites

• A Discord application and bot (from the Discord Developer Portal)

Setup

1. Go to Discord Developer Portal.
2. Click "New Application" and name it.
3. Go to the Bot section and click "Add Bot".
4. Copy the bot token.
5. Enable Message Content Intent required to read message content.
6. Set the environment variable:

export DISCORD_BOT_TOKEN=MTIzNDU2Nzg5.ABCDEF.ghijklmnop

7. Add to config:

[channels.discord]
bot_token_env = "DISCORD_BOT_TOKEN"
default_agent = "coder"

8. Restart the daemon.

How It Works

The Discord adapter connects to the Discord Gateway via WebSocket (v10) and routes messages to the configured agent.

Slack

Prerequisites

• A Slack app with Socket Mode enabled

Setup

1. Go to Slack API and click "Create New App" > "From Scratch".
2. Copy the token.
3. Set the environment variables:

export SLACK_APP_TOKEN=xapp-1-...
export SLACK_BOT_TOKEN=xoxb-...

4. Add to config:

[channels.slack]
bot_token_env = "SLACK_BOT_TOKEN"
app_token_env = "SLACK_APP_TOKEN"
default_agent = "ops"

5. Restart the daemon.

How It Works

The Slack adapter uses Socket Mode to receive events and routes them to the configured agent.