Search conversations by message content
Searches conversation messages by content and falls back to a case-insensitive match when no ranked results are found.
/api/conversations/searchSearches conversation messages by content and falls back to a case-insensitive match when no ranked results are found. Results are ordered by relevance, include matching snippets, and are paginated within the documented search window. Search offsets are capped to protect the ranked query; refine the search phrase when the requested offset exceeds the documented search window.
Endpoint requirements
- Credentials
- Use either
Bearer API keyorX-API-Key header. - Required scope
conversations:read- Module
- Automation & API module (
api_access) - Organization and site scope
- Requests can access only resources owned by the authenticated organization. The API does not reveal whether an identifier belongs to another organization or site.
- Retry behavior
- Safe read: You can safely retry this read-only request after a transport failure. It does not change resource state.
Query Parameters
Full-text search query.
1 <= length <= 500Maximum number of search results to return.
1 <= value <= 5020Zero-based number of search results to skip.
0 <= value <= 90071992547409910DA-4035: alternative to X-Project-Id: all for the all-sites aggregate scope on read-only aggregate routes. Only all is accepted; any other value returns PROJECT_SCOPE_NOT_SUPPORTED (400).
Value in
- "all"
Header Parameters
DA-4035: selects the active site (project) for project-scoped operations. Accepts a site UUID, or the literal all to aggregate across every accessible site on read-only aggregate routes (single-or-all-read). Mutations and single-site routes reject all. Path-owned project routes (/api/projects/{projectId}/…) resolve the site from the URL and ignore this header. Conflicting selectors across header/query/body/path return PROJECT_CONTEXT_CONFLICT (400); an archived site returns PROJECT_ARCHIVED (410).
Response Body
application/json
application/json
application/json
application/json
application/json
application/json
application/json
application/json
application/json
application/json
{ "data": [ { "id": "11111111-1111-4111-8111-111111111111", "organizationId": "22222222-2222-4222-8222-222222222222", "projectId": "11111111-1111-4111-8111-111111111111", "projectName": "Example name", "visitorId": "33333333-3333-4333-8333-333333333333", "operatorId": "44444444-4444-4444-8444-444444444444", "status": "open", "priority": "normal", "source": "api", "aiEnabled": true, "locale": "en", "localeSource": "organization_default", "localeLocked": false, "localeUpdatedAt": null, "title": "Billing question", "messageCount": 2, "hasUnreadMessages": true, "tags": [ "billing" ], "customFields": {}, "languageConfidence": 1, "languageSource": "manual", "translationActive": true, "translationTargetLanguage": "example", "createdAt": "2026-08-03T08:30:00.000Z", "updatedAt": "2026-08-03T08:35:00.000Z" } ], "pagination": { "totalItems": 1, "page": 1, "pageSize": 20, "hasNextPage": false, "hasPrevPage": false }}Request examples
curl
curl --request GET \
'https://api.convor.io/api/conversations/search?q=payment+issue' \
--header 'Accept: application/json' \
--header 'Authorization: Bearer convor_sk_...'JavaScript
const response = await fetch("https://api.convor.io/api/conversations/search?q=payment+issue", {
method: "GET",
headers: {
"Accept": "application/json",
"Authorization": "Bearer convor_sk_..."
},
});
const contentType = response.headers.get("content-type") ?? "";
const data = response.status === 204
? null
: contentType.includes("json")
? await response.json()
: contentType.startsWith("text/")
? await response.text()
: await response.blob();
if (!response.ok) {
const message = typeof data === "object" && data !== null
&& "error" in data && typeof data.error === "object"
&& data.error !== null && "message" in data.error
? String(data.error.message)
: "Convor API request failed (" + response.status + ")";
throw new Error(message);
}
console.log(data);Was this page helpful?