Documentation
Comms
comms.list_conversations
REST
ShippingPOST /api/v1/comms/list_conversationsMCP tool
Livecomms.list_conversationsExposed on: REST API · Shop MCP server · Claude connector · Dash (in-app copilot) · Zapier. Part of the Comms domain.
Operating contract
The inbox itself: one row per thread, most recently active first, narrowed by status, channel, customer_id, who it is assigned to, and whether it is waiting on a reply.
unread IS A HEURISTIC, NOT A FLAG SOMEBODY SET. It is true when the customer's last message landed after our last outbound — i.e. the ball is in the shop's court. That is the question 'what needs answering' actually means, and it is why unread_only is the filter a morning triage should use rather than status.
status is the SHOP'S disposition: open is live, closed is dealt with (Quo calls this 'done'), snoozed is parked. A snoozed thread reopens on the next inbound, so a thread being snoozed today says nothing about tomorrow.
Each result's id is the conversation id that comms.get_conversation, comms.list_messages, comms.update_conversation, comms.schedule_message and tasks.create take. customer_id is what customers.get takes; it is null on a thread from a number nobody has matched to a contact yet, and comms.resolve_contact is how to find out who that is.
TRUNCATION: total is the exact number of threads matching the filter and omitted how many are not in this response. A busy shop's inbox runs to thousands of threads, so 'how many are waiting on us' must come from total. When omitted is not 0, call again with offset set to next_offset.
Who may call it
- Permission
inbox.accessThe caller must hold Inbox at the ACT level. A read-only dashboard grant on the same section is refused.- Plan
- Every planNo plan gate. Available on every Service VIN plan.
- Retries
naturalNaturally idempotent — running it twice leaves the same world as running it once. A retrying integration needs no key.- Rate class
readCounted against the read budget — the widest of the four.
Partial answers
omitted is how many matching records are not in the response, and 0 is a real answer meaning you have all of them. Continue with next_offset.
Do not summarise from a truncated response
Input
| Field | Type | Description |
|---|---|---|
| status | string | Only threads at this disposition. Omit for all three. One of:open, snoozed, closed |
| channel | string | Only threads on this channel. `call` is a thread a phone call opened. One of:sms, email, call, webchat, whatsapp |
| customer_id | stringuuid | Only this customer's threads, as returned by customers.list or comms.resolve_contact. |
| staff_id | stringuuid | Only threads assigned to this teammate, as returned by tasks.assignees. |
| unassigned_only | boolean | True for threads nobody has taken. Ignored when staff_id is given. Default:false |
| unread_only | boolean | True for threads whose last word was the customer's — the ones waiting on a reply. Default:false |
| limit | integer1–100 | How many threads to return, 1-100. Defaults to 25. Default:25 |
| offset | integermin 0 | How many to skip. Pass the previous response's next_offset to continue. Default:0 |
Output
| Field | Type | Description |
|---|---|---|
| items | object[] | — |
| items[].id | string | — |
| items[].customer_id | string | null | — |
| items[].channel | string | — |
| items[].status | string | — |
| items[].subject | string | null | — |
| items[].contact_number | string | null | — |
| items[].assigned_to | string | null | — |
| items[].pinned | boolean | — |
| items[].unread | boolean | — |
| items[].is_ai_managed | boolean | — |
| items[].last_message_at | string | null | — |
| items[].last_inbound_at | string | null | — |
| items[].last_outbound_at | string | null | — |
| items[].last_human_outbound_at | string | null | — |
| items[].created_at | string | — |
| total | number | — |
| omitted | number | — |
| next_offset | number | null | — |
Examples
Built from this capability's own schema — required fields and the ones carrying a default, and nothing invented. Paste one and it validates.
export SERVICEVIN_API_KEY=svk_live_…
curl -X POST https://www.servicevin.com/api/v1/comms/list_conversations \
-H "Authorization: Bearer $SERVICEVIN_API_KEY" \
-H "Content-Type: application/json" \
-d '{"unassigned_only":false,"unread_only":false,"limit":25,"offset":0}'const res = await fetch("https://www.servicevin.com/api/v1/comms/list_conversations", {
method: "POST",
headers: {
Authorization: `Bearer ${process.env.SERVICEVIN_API_KEY}`,
"Content-Type": "application/json",
},
body: JSON.stringify({
"unassigned_only": false,
"unread_only": false,
"limit": 25,
"offset": 0
}),
});
// Success and failure are both envelopes. Switch on error.code, never
// on error.message — the codes are stable, the messages are for people.
const payload = await res.json();
if (!res.ok) throw new Error(payload.error.code);
const data = payload.data;import os, requests
res = requests.post(
"https://www.servicevin.com/api/v1/comms/list_conversations",
headers={"Authorization": f"Bearer {os.environ['SERVICEVIN_API_KEY']}"},
json={
"unassigned_only": False,
"unread_only": False,
"limit": 25,
"offset": 0
},
timeout=30,
)
payload = res.json()
if not res.ok:
raise RuntimeError(payload["error"]["code"])
data = payload["data"]{
"jsonrpc": "2.0",
"id": 1,
"method": "tools/call",
"params": {
"name": "comms.list_conversations",
"arguments": {
"unassigned_only": false,
"unread_only": false,
"limit": 25,
"offset": 0
}
}
}Refusals
The four gates run in this order on every surface, and the order is not arbitrary — see Authentication.
| Status | Code | When |
|---|---|---|
| 404 | not_found | The id is unknown, or the feature is not enabled for this account. Deliberately the same answer for both. |
| 403 | forbidden | This login does not hold inbox.access. |
| 422 | validation_error | An argument was wrong. The message names the field. |
| 429 | rate_limited | Too many read calls. Back off and retry. |
| 500 | internal_error | Something failed on our side. Nothing was changed. |