Add a Capability to a Specialist

Without capabilities, an Agent Stack can only talk. With them, it can act — call your order API, process a cancellation, look up a customer record, fire a webhook. Capabilities are attached per specialist, so the order tracker can call the order-lookup API while the billing agent can only read invoices.

Before you start

• The capability must already exist in Settings → Capabilities. If not, create one first.
• Decide the security level — what authentication state must the customer be in for this capability to fire?
• L0 — always available, even to anonymous callers (e.g. order-status lookup with the order ID).
• L1 — after the customer claims an identity (e.g. they typed their email).
• L2 — after OTP verification.
• L3 — after full authentication.

Pick the lowest level that’s safe. L0 is fine for read-only public data; L3 is required for anything that changes account state.

For voice stacks and phone calls, only L0 capabilities are available to the agent. L1–L3 assignments still apply in other channels, but they are withheld from voice.

Steps

1. Open Settings → Agent Stacks → [stack] → Orchestrator.
2. Click the specialist node that should get the new capability.
3. Click the Capabilities sub-tab.
4. Click + Add capability.
5. Pick the capability from the list. (You’ll only see capabilities your tenant has defined.)
6. Set the security level for this assignment:
   Leave it as Inherit to use the capability’s default security level (set when the capability itself was created).
   Override to a specific L0/L1/L2/L3 to use a different level for this specialist.
7. Set the approval mode:
   Auto — the AI calls the capability without human review.
   Human — the AI prepares the call but a human approves before it fires.
8. Save.

Verify it worked

• The Capabilities sub-tab now shows the new capability in the list with the configured security level and approval mode.
• The specialist’s slide-out summary card (top of the panel) shows a non-zero Capabilities count.
• Open the Testing tab and ask the specialist a question that should trigger the capability. The reply should include the action result (e.g. order status, refund confirmation).

When to override the default security level

Capabilities have a default security level set at the capability itself. Override on a per-specialist basis when:

• A specialist serves higher-risk customers and you want to require stronger auth (e.g. an enterprise specialist requires L3 for actions that another stack runs at L1).
• A specialist serves lower-risk customers and you want to relax auth (e.g. internal-only stack runs everything at L0).

When in doubt, leave it as Inherit — the capability’s default is usually correct.

Troubleshooting

• Symptom: A “warning” icon appears next to a capability in the list. Fix: that’s the L0-on-action warning — actions (capabilities that change state, not read state) shouldn’t generally run at L0. Bump the security level to L1+ unless you have a deliberate reason.
• Symptom: The capability never fires in test. Fix: confirm the test session’s auth level meets the capability’s effective level. Open the test panel, click Verify identity, and bring the auth level up. If you’re testing over voice, confirm the capability is L0.
• Symptom: The capability fires but the AI doesn’t reference its result. Fix: check the specialist’s Instructions — sometimes a poorly worded instruction tells the AI to ignore certain capability outputs. Also check the capability’s response format in Settings → Capabilities; if the AI can’t parse the response, it won’t use it.
• Symptom: Approval mode is Human but no approval ever comes through. Fix: approvals appear in the conversation as a pending action. The handler must approve from the conversation detail panel; configure who can approve via Roles & Permissions.

See also

• What is a specialist agent?
• What are Capabilities?
• Add a specialist agent