The intercom API reference.
The intercom API reference.
Conversations are how you can communicate with users in Intercom. They are created when a contact replies to an outbound message, or when one admin directly sends a message to a single contact.
The last time a Contact responded to an Admin. In other words, the time a customer started waiting for a response. Set to null if last reply is from an Admin.
If set this is the time in the future when this conversation will be marked as open. i.e. it will be in a snoozed state until this time. i.e. it will be in a snoozed state until this time.
If marked as priority, it will return priority or else not_priority.
The id of the admin assigned to the conversation. If it's not assigned to an admin it will return null.
The id of the team assigned to the conversation. If it's not assigned to a team it will return null.
The Conversation Rating object which contains information on the rating and/or remark added by a Contact and the Admin assigned to the conversation.
The Conversation Part that originated this conversation, which can be Contact, Admin, Campaign, Automated or Operator initiated.
The list of contacts (users or leads) involved in this conversation. This will only contain one customer unless more were added via the group conversation feature.
The list of teammates who participated in the conversation (wrote at least one conversation part).
An object containing the different custom attributes associated to the conversation as key-value pairs. For relationship attributes the value will be a list of custom object instance models.
An object containing information on the first users message. For a contact initiated message this will represent the users original message.
The SLA Applied object contains the details for which SLA has been applied to this conversation. Important: if there are any canceled sla_events for the conversation - meaning an SLA has been manually removed from a conversation, the sla_status will always be returned as null.
A Statistics object containing all information required for reporting, with timestamps and calculated metrics.
A list of Conversation Part objects for each part message in the conversation. This is only returned when Retrieving a Conversation, and ignored when Listing all Conversations. There is a limit of 500 parts.
An object containing metadata about linked conversations and linked tickets. Up to 1000 can be returned.
Indicates whether the AI Agent participated in the conversation.
{ "type": "conversation", "id": "1295", "title": "Conversation Title", "created_at": 1663597223, "updated_at": 1663597260, "waiting_since": 1663597260, "snoozed_until": 1663597260, "open": true, "state": "open", "read": true, "priority": "priority", "admin_assignee_id": 0, "team_assignee_id": "5017691", "tags": { "type": "tag.list", "tags": [ … ] }, "conversation_rating": { "rating": 5, "remark": "", "created_at": 1671028894, "contact": { … }, "teammate": { … } }, "source": { "type": "conversation", "id": "3", "delivered_as": "operator_initiated", "subject": "", "body": "<p>Hey there!</p>", "author": { … }, "attachments": [ … ], "url": null, "redacted": false }, "contacts": { "type": "contact.list", "contacts": [ … ] }, "teammates": { "type": "admin.list", "teammates": [ … ] }, "custom_attributes": { "property1": "string", "property2": "string" }, "first_contact_reply": { "created_at": 1663597223, "type": "conversation", "url": "https://developers.intercom.com/" }, "sla_applied": { "type": "conversation_sla_summary", "sla_name": "", "sla_status": "hit" }, "statistics": { "type": "conversation_statistics", "time_to_assignment": 2310, "time_to_admin_reply": 2310, "time_to_first_close": 2310, "time_to_last_close": 2310, "median_time_to_reply": 2310, "first_contact_reply_at": 1663597233, "first_assignment_at": 1663597233, "first_admin_reply_at": 1663597233, "first_close_at": 1663597233, "last_assignment_at": 1663597233, "last_assignment_admin_reply_at": 1663597233, "last_contact_reply_at": 1663597233, "last_admin_reply_at": 1663597233, "last_close_at": 1663597233, "last_closed_by_id": "c3po", "count_reopens": 1, "count_assignments": 1, "count_conversation_parts": 1 }, "conversation_parts": { "type": "conversation_part.list", "conversation_parts": [ … ], "total_count": 2 }, "linked_objects": { "type": "list", "total_count": 100, "has_more": false, "data": [ … ] }, "ai_agent_participated": true, "ai_agent": { "source_type": "workflow", "source_title": "My AI Workflow", "last_answer_type": "ai_answer", "resolution_state": "assumed_resolution", "rating": 4, "rating_remark": "Very helpful!", "content_sources": { … } } }
curl -i -X POST \
'https://api.intercom.io/conversations/{conversation_id}/tags' \
-H 'Authorization: Bearer <YOUR_TOKEN_HERE>' \
-H 'Content-Type: application/json' \
-H 'Intercom-Version: 2.11' \
-d '{
"id": 99,
"admin_id": 991267526
}'
{ "type": "tag", "id": "99", "name": "Manual tag" }
curl -i -X DELETE \
'https://api.intercom.io/conversations/{conversation_id}/tags/{id}' \
-H 'Authorization: Bearer <YOUR_TOKEN_HERE>' \
-H 'Content-Type: application/json' \
-H 'Intercom-Version: 2.11' \
-d '{
"admin_id": 991267530
}'
{ "type": "tag", "id": "102", "name": "Manual tag" }
You can fetch a list of all conversations.
You can optionally request the result page size and the cursor to start after to fetch the result.
You can use pagination to limit the number of results returned. The default is 20
results per page. See the pagination section for more details on how to use the starting_after
param.
curl -i -X GET \
'https://api.intercom.io/conversations?per_page=20&starting_after=string' \
-H 'Authorization: Bearer <YOUR_TOKEN_HERE>' \
-H 'Intercom-Version: 2.11'