Building with HiveMind
HiveMind is the AI copilot built into your dashboard. Instead of clicking through builders, you describe an outcome and HiveMind assembles it — bots, integrations, and workflows.
What it can do
- Build a bot: "Make a bot that triages support emails and drafts replies."
- Wire an integration: "Connect my Notion workspace so a bot can create pages." It runs the Architect — interviewing you, researching the API, and scaffolding the integration.
- Design a workflow: "When a lead comes in, score it, and if it's hot, alert Slack." It returns a workflow card you can open in the Orchestrator.
How to get the best results
- State the goal, not the steps. "Summarize new GitHub issues every morning and post to Slack" beats a vague "help with GitHub."
- Name your own bots/apps to have them wired in: "Have my Triage bot handle it."
- Iterate in chat. Ask for changes — "use a cheaper model on the classify step" — and it adjusts.
Memory-aware
HiveMind knows what you've already connected. If your Slack is connected, it won't ask you to connect it again — it just uses it.
From chat to canvas
When HiveMind proposes a workflow, click Open in builder to drop it on the Orchestrator canvas, fine-tune any node, and save. Bots it builds land in your bot list, ready to test and deploy.
Next steps
- Build your first bot — the manual path, for full control.
- Build & schedule workflows — what HiveMind generates under the hood.
Use the test pane to iterate quickly. Every change you make is live — no need to save first.
API keys are shown only once. Store them securely and never commit them to version control.
Test your bot with edge cases before deployment. Try empty inputs, long messages, and special characters.
Chain multiple specialized bots in a workflow for better results than one general-purpose bot.