Skip to content

Support

AI agents that use Intercom

UtilitySpaces gives an AI agent scoped access to Intercom through its own encrypted connection, not your password. The agent can call 5 named Intercom tools, and the one that writes waits for a person's approval unless an admin has set the connection to run without asking. Every call is written to a hash-chained audit log with its arguments and its result.

What an agent can do in Intercom

Look up contacts and conversations in Intercom and reply to conversations as an admin. An agent calls these by name, and matching is exact.

Every tool a Intercom connection exposes, and whether it writes, which waits for approval by default
ToolWhat it doesWaits for approval by default
Get current adminCheck the token and identify the authenticated admin and app.No — read only
List contactsList contacts (users and leads).No — read only
Search contactsSearch contacts by field query (email, name, custom attributes).No — read only
List conversationsList conversations in the workspace.No — read only
Reply to conversationPost an admin reply or note into a conversation.Yes — it writes

4 read tools, 1 write tool. Generated from the connector catalogue the product runs on, so it cannot list a tool that does not exist.

What teams hand an agent in Intercom

Teams hand an agent the context-gathering in Intercom: finding a contact by email or custom attribute with Search contacts, listing their conversations, and assembling what the person has already been told. Reply to conversation is the one write, covering both admin replies and internal notes, and it waits for a person's approval unless an admin has set this connection to run without asking.

  • Look up a contact's custom attributes when a conversation arrives, so the reply is written with their account details in view.
  • Keep a list of unanswered conversations current by listing them on a schedule and flagging the oldest in the team inbox.
  • Post an internal note on a conversation when related context turns up elsewhere, such as a matching support ticket (after a person approves it)
Before you let it write, or keep out: Reply to conversation can post either a reply the contact sees or a note only admins see; one write policy covers both, so read what each approval will actually post: a call meant as a note can go out as a reply.

What waits for a person

Reads run on their own. These do not — each one waits in Approvals for an owner or admin unless an admin has set this connection to run writes without asking, and every call toIntercom, approved, rejected or unattended, is logged.

  • Reply to conversation POST

What connecting Intercom takes

Intercom uses an API key you create in Intercom and paste in once. A workspace admin sets the connection up; the fields are:

  • Access Token

Credentials are encrypted and bound to your workspace, never read back, and every outbound call passes one guarded fetch — the connector security posture sets out exactly how, and the security page covers the platform around it.

Where Intercom fits

Give an agent Intercom work by writing a message in the inbox, or put its tools in a scheduled flow. What the run cost is on the cost breakdown.

Other support integrations: Freshdesk, Zendesk. Or see every application agents can connect to.

See it doing your actual work

Open a demo workspace, connect one app, and give an agent a real job. No card, no sales call.

We reply to every message within one business day.

Try the demo