Skip to content
DigitalBrain Wiki edited this page Jul 16, 2026 · 2 revisions

Demos

Three short demos. Order: A → B → C. Full engineering runbooks live with the team; this is the stakeholder version.

Always use the same account/workspace for app and agent.


A — Enter the workspace (~3 min)

  1. Show Sign in to DigitalBrain.
  2. Sign in.
  3. Point at destinations: Chat · Features · Connections · Activity.

Say: “This is a workspace of live surfaces — not four mini-apps with their own truth.”

Close: Login puts you in one brain; the app only projects it.


B — Agent + live app (~5 min)

  1. Stay signed in; open Chat (and Activity if useful).
  2. Agent joins DigitalBrain MCP on the same workspace.
  3. Agent sends a clear probe message (unique phrase).
  4. App updates without restarting.
  5. Optional: agent retries the same command id → no double work.

Say: “The agent is another client of the same Chat — not a second brain.”

Close: When the agent acts, you see it live.


C — Connect and control (~5–8 min)

  1. Open Connections.
  2. Connect Google or Salesforce → Healthy.
  3. Show Unlocked capabilities.
  4. If you can: prepare an outside change → Approve or Reject → show Activity.
  5. If you cannot complete a real approval: stop at health + unlocks and say so honestly.

Say: “Connect once, unlock abilities, approve what touches the world — no agent packs.”

Close: Connect → unlock → approve → done.


Do not say

  • Agent marketplace / Gmail agent pack
  • Paste any MCP URL
  • UI brain vs agent brain
  • Swarm OS

Pass / fail (simple)

Demo Pass
A Four destinations after sign-in; “one workspace” story
B Agent work visible in app; same workspace
C Healthy connection + unlocks (and approval if available)

In one line: Enter · agent lights the app · connect and approve.

Back: Home · How it works

Clone this wiki locally