Skip to main content

AI Employees Command Centre

Command Centre (/ai-employees/command-center) is the HUD for AI workforce operations—agent pipeline, status chips, queues, and alerts across all deployed agents. Use it during marketing sends, calling campaigns, or inbox spikes to spot failing credentials before customers notice.

Video

Video coming soon for this topic

Who can do this​

AI ops users and Admins.

Before you begin​

  • Multiple live agents
  • Defined on-call owner for credential fixes

What you'll accomplish​

  • Fleet health reviewed
  • Failing agent diagnosed and retested

Step-by-step procedure​

Daily ops​

  1. Sidebar → AI Employees → Command Centre.
  2. Scan workforce pipeline (avatar rings + status chips).
  3. Open agents flagged degraded or offline.
  4. Jump to Tools / Credentials; run connection test.
  5. Re-run sample chat or test call.

During campaigns​

  1. Watch concurrent voice/chat load while Calling Campaigns run.
  2. Pause campaigns if Command Centre shows systemic tool failures.
  3. Coordinate with Credits and usage for spend spikes.

Reporting tie-in​

Export agent performance notes to leadership; deep metrics also appear on agent Activity and AI usage analytics.

Extended guidance​

Command Centre is not a replacement for Monitor Smart Flow executions—use both during automation incidents. During incidents, snapshot HUD metrics before fixing credentials for postmortems. Pair with AI Usage Analytics to explain credit burn spikes.

Practice scenarios​

Simulate credential failure on one agent, observe Command Centre alert, fix, confirm green status. Document response time from alert to successful test chat for your ops runbook.

Sandbox validation​

Repeat the procedure in a test workspace or with two internal colleagues before customer-facing launches. Document who approved each step (marketing, legal, ops). Screenshot Verify your work outcomes for audit trails. When something fails, capture the URL, role, timestamp, and exact UI label clicked—Support resolves tickets faster with that context. Re-read See also links after your first successful run; adjacent topics often cover edge cases you have not hit yet.

Team rollout checklist​

Assign a single owner for this workflow per workspace. Schedule a 30-minute handoff training with screen share. Add the See also links to your internal Notion or wiki. Re-verify after each major product release (labels move occasionally). Pair junior staff with a senior reviewer for the first live execution. Log lessons learned in your team retro—especially permission surprises and integration timeouts.

Keep learning​

Bookmark this page and the Product module map. After major releases, re-walk Step-by-step procedure—UI labels move occasionally. Share feedback with your workspace Admin so Help stays aligned with how your team actually works.

Verify your work​

  • Alert count returns to baseline
  • Test conversation succeeds post-fix
  • No orphaned agents stuck offline

Tips​

tip

Assign one Admin to weekly Command Centre review— catches credential expiry early.

Troubleshooting​

| Empty dashboard | No agents deployed yet | | Status flaps | Unstable webhook or rate limit—check Activity | | Cannot open page | AI Employees permission missing |

See also​