Skip to content
QuantCDN

Creating your first Slack bot

This guide walks you through creating a Slack bot powered by Quant AI agents. By the end, your team will be able to interact with AI directly in Slack.

Prerequisites

Section titled “Prerequisites”

Before creating a Slack bot, you need:

  • At least one AI agent configured in your organisation
  • The manage_ai_agents permission (Organisation Owner or Admin role)
  • Access to a Slack workspace where you can install apps (for the Quant-managed path) or create apps (for the BYO path)
Section titled “Navigate to Slack Bots”

In the dashboard sidebar, click Slack Bots under the AI section, then click Create Bot to open the setup wizard.

Setup wizard

Section titled “Setup wizard”

The bot creation wizard has three steps.

Step 1: Choose setup type

Section titled “Step 1: Choose setup type”

Select how you want to connect your bot to Slack:

  • Quant-managed (recommended) — Use the shared Quant AI Slack app. One click to install via OAuth. Bot appears as “Quant AI” in your workspace.
  • Bring Your Own (BYO) — Create your own Slack app with a custom name and icon. You control the app identity and slash command names.

See Setup types for a detailed comparison.

Step 2: Configure

Section titled “Step 2: Configure”

Configure your bot’s behaviour:

Bot name — A display name for identifying the bot in the dashboard.

Sub-agents — Select the agents your bot can delegate to. You can select multiple agents — the bot automatically creates a router agent that directs questions to the right specialist based on the conversation.

If you select a single agent, the bot delegates all questions to that agent. If you select multiple agents, the bot evaluates each question and routes it to the most relevant specialist.

System prompt — Custom instructions for your bot. If sub-agents are also selected, the routing instructions are appended automatically.

Tools, Skills, Knowledge Base — Expand these collapsible sections to assign capabilities directly to the bot’s agent, just like configuring any other agent.

Session TTL — How long conversation history is retained per thread (1—90 days, default 14).

Keyword monitoring — Optionally enable keyword triggers so the bot responds when specific words appear in allowed channels.

Advanced configuration — Expand this section to set model, temperature, max tokens, guardrail presets, and filter policies.

Step 3: Connect

Section titled “Step 3: Connect”

The final step depends on your setup type:

Quant-managed: Click Add to Slack to start the OAuth flow. You’ll be redirected to Slack to authorise the bot in your workspace, then back to the dashboard.

BYO: Copy the webhook URL displayed and configure it in your Slack app settings. Then paste your Bot Token and Signing Secret into the form fields.

Verify the bot is working

Section titled “Verify the bot is working”

Once connected, test the bot in Slack:

  1. Open a channel where the bot has access (or send a direct message)
  2. @mention the bot with a question, for example: @Quant AI What can you help me with?
  3. The bot posts a progress message that updates in real-time, then displays the final answer

What happens behind the scenes

Section titled “What happens behind the scenes”

When you save the bot configuration, the portal:

  1. Creates an agent to power the bot (with your selected tools, skills, knowledge bases, and sub-agent routing)
  2. Creates the bot record linked to that agent
  3. Stores your Slack credentials (KMS-encrypted, write-only)
  4. Activates the event webhook so Slack messages reach your bot

The bot’s agent appears in your AI Agents list with a “Slack Bot” badge for easy identification.

Next steps

Section titled “Next steps”