Bot Interfaces

DjinnBot agents can interact through multiple interfaces. Currently Slack and the built-in dashboard chat are supported, with more platforms planned.

Current Interfaces

Dashboard Chat

The built-in chat interface at http://localhost:3000/chat requires no additional setup. Features:

• Select any agent and model
• Full tool access (code execution, file operations, web research)
• Persistent chat history
• Real-time streaming responses
• Supports onboarding and project-context sessions

This is the primary interface for users who don’t use Slack.

Slack

Each agent gets its own Slack bot via Socket Mode. See Slack Bot Setup for configuration.

Features:

• Per-agent bot identity (name, avatar)
• Pipeline threads (watch agents collaborate)
• Direct mentions and DMs
• Active/passive thread participation

Planned Interfaces

Discord

Discord bot support is on the roadmap. The architecture mirrors Slack — each agent gets its own bot account, pipeline runs create threads, and agents respond to mentions.

Microsoft Teams

Teams integration is planned for enterprise environments that standardize on the Microsoft ecosystem.

Custom Webhooks

A generic webhook interface will allow integration with any chat platform or custom application. Send messages in, receive agent responses out, via simple HTTP.

API-Only

The REST API and SSE streaming already allow building custom frontends. Any application that can make HTTP requests can interact with DjinnBot agents.

Architecture for New Interfaces

Adding a new chat interface involves:

1. Bridge service — connects to the external platform’s API (similar to packages/slack/)
2. Event routing — maps platform events to DjinnBot’s event bus
3. Per-agent identity — manages bot accounts/tokens for each agent
4. Thread mapping — links platform threads to pipeline runs

The engine’s event-driven architecture makes this straightforward — new interfaces subscribe to Redis events and publish commands back.

Contributing an Interface

If you want to add support for a new platform, the Slack package (packages/slack/) is the reference implementation. The key files:

• slack-bridge.ts — routes events between Slack and the engine
• agent-slack-runtime.ts — manages per-agent Socket Mode connections
• thread-manager.ts — maps runs to Slack threads
• slack-streamer.ts — streams agent output to Slack messages

A new interface would implement the same patterns for a different platform.