> ## Documentation Index > Fetch the complete documentation index at: https://docs.gleap.ai/llms.txt > Use this file to discover all available pages before exploring further. # Find messages by query > Get messages by query with support for filtering and pagination. **Filtering:** - Filter by ticket: `ticket=507f1f77bcf86cd799439011`. Any identifier `GET /tickets/{ticketId}` accepts works here too — the human-facing ticket number shown in the dashboard (`ticket=143790`), the 24-char ObjectId, or the shareToken. - Filter by type: `type=TEXT` or `type=TEXT,NOTE` (comma-separated values match any; bracket syntax such as `type[$in]=...` is NOT supported) - Filter by bot messages: `bot=true` or `bot=false` - Filter by date range: `createdAt>=2024-01-01&createdAt<=2024-12-31` **Ordering:** Messages are always returned oldest-first (ascending `createdAt`); the sort order is not configurable. **Pagination:** - Cursor mode (recommended): pass `paginated=true` to receive `{ items, hasMore, nextBefore }`; pass `before=` (the previous `nextBefore`) to fetch the next older page. `skip` is ignored in this mode; `limit` defaults to 30 and is clamped to 1-100. - Legacy mode (no `paginated`): returns a flat array; `limit` defaults to 1000 and values above 100 are clamped to 100; `skip` is applied. **Exporting a full conversation:** Filter by `ticket` rather than by `bot`. `bot=true` selects messages exchanged with the AI bot, so on a conversation that was handed over to a human it omits the agent's replies — `ticket={ticketId}` returns the complete transcript. Thread replies are not part of the main feed either; fetch them per message with `GET /messages/{messageId}/thread`. Which conversations still exist is governed by retention — AI-only (`type=BOT`) conversations are deleted about 33 days after they are created, see GET /tickets. **Ticket history:** Since August 2026 this endpoint returns conversation messages only. Audit entries — `FEEDBACK_UPDATED` (status, assignee, team, priority, type, tag, title changes, …) and workflow/system entries — are no longer part of the response; filtering by those types returns an empty result. Fetch them from `GET /tickets/{ticketId}/history`, which returns them in the same message-compatible shape. **Translation:** - `language=es` returns messages translated into the given language. Requires the project setting 'translate customer messages'; only conversation messages (`TEXT`, `USER_TEXT`, `NOTE`, `BOT`, `BOT_REPLY`, `SHARED_COMMENT`) are translated, and messages already in the target language are returned unchanged. ## OpenAPI ````yaml https://api.gleap.io/api-docs.json get /messages openapi: 3.0.0 info: title: gleap-server version: 14.0.0 contact: {} servers: - url: https://api.gleap.io/v3 security: [] paths: /messages: get: tags: - Messages summary: Find messages by query description: >- Get messages by query with support for filtering and pagination. **Filtering:** - Filter by ticket: `ticket=507f1f77bcf86cd799439011`. Any identifier `GET /tickets/{ticketId}` accepts works here too — the human-facing ticket number shown in the dashboard (`ticket=143790`), the 24-char ObjectId, or the shareToken. - Filter by type: `type=TEXT` or `type=TEXT,NOTE` (comma-separated values match any; bracket syntax such as `type[$in]=...` is NOT supported) - Filter by bot messages: `bot=true` or `bot=false` - Filter by date range: `createdAt>=2024-01-01&createdAt<=2024-12-31` **Ordering:** Messages are always returned oldest-first (ascending `createdAt`); the sort order is not configurable. **Pagination:** - Cursor mode (recommended): pass `paginated=true` to receive `{ items, hasMore, nextBefore }`; pass `before=` (the previous `nextBefore`) to fetch the next older page. `skip` is ignored in this mode; `limit` defaults to 30 and is clamped to 1-100. - Legacy mode (no `paginated`): returns a flat array; `limit` defaults to 1000 and values above 100 are clamped to 100; `skip` is applied. **Exporting a full conversation:** Filter by `ticket` rather than by `bot`. `bot=true` selects messages exchanged with the AI bot, so on a conversation that was handed over to a human it omits the agent's replies — `ticket={ticketId}` returns the complete transcript. Thread replies are not part of the main feed either; fetch them per message with `GET /messages/{messageId}/thread`. Which conversations still exist is governed by retention — AI-only (`type=BOT`) conversations are deleted about 33 days after they are created, see GET /tickets. **Ticket history:** Since August 2026 this endpoint returns conversation messages only. Audit entries — `FEEDBACK_UPDATED` (status, assignee, team, priority, type, tag, title changes, …) and workflow/system entries — are no longer part of the response; filtering by those types returns an empty result. Fetch them from `GET /tickets/{ticketId}/history`, which returns them in the same message-compatible shape. **Translation:** - `language=es` returns messages translated into the given language. Requires the project setting 'translate customer messages'; only conversation messages (`TEXT`, `USER_TEXT`, `NOTE`, `BOT`, `BOT_REPLY`, `SHARED_COMMENT`) are translated, and messages already in the target language are returned unchanged. operationId: GetMessages parameters: - in: header name: project required: true schema: type: string - description: >- - Set to `true` for cursor mode: returns `{ items, hasMore, nextBefore }` in: query name: paginated required: false schema: type: boolean - description: >- - Cursor for the next older page (the previous response's `nextBefore`); only with `paginated=true` in: query name: before required: false schema: type: string - description: >- - Maximum number of messages to return. Cursor mode: default 30, clamped to 1-100. Legacy mode: default 1000, values above 100 are clamped to 100. in: query name: limit required: false schema: type: number format: double example: '50' - description: >- - Deprecated, accepted and ignored. Ticket-history entries are served by `GET /tickets/{ticketId}/history`. in: query name: showHistory required: false schema: type: boolean deprecated: true responses: '200': description: Ok content: application/json: schema: anyOf: - items: {} type: array - properties: nextBefore: {} hasMore: type: boolean items: items: {} type: array required: - nextBefore - hasMore - items type: object examples: Example 1: value: - id: 507f1f77bcf86cd799439015 ticket: 507f1f77bcf86cd799439011 type: TEXT data: content: type: doc content: - type: paragraph content: - type: text text: This is a sample message content. comment: This is a sample message content. bot: false user: id: 507f1f77bcf86cd799439012 email: support@example.com firstName: John lastName: Doe session: _id: 507f1f77bcf86cd799439013 id: 507f1f77bcf86cd799439013 email: user@example.com createdAt: '2024-01-15T10:30:00.000Z' updatedAt: '2024-01-15T10:30:00.000Z' index: 0 - id: 507f1f77bcf86cd799439016 ticket: 507f1f77bcf86cd799439011 type: TEXT data: content: type: doc content: - type: paragraph content: - type: text text: Thank you for the update! comment: Thank you for the update! bot: false user: id: 507f1f77bcf86cd799439017 email: user@example.com firstName: Jane lastName: Smith session: _id: 507f1f77bcf86cd799439013 id: 507f1f77bcf86cd799439013 email: user@example.com createdAt: '2024-01-15T11:00:00.000Z' updatedAt: '2024-01-15T11:00:00.000Z' index: 1 security: - jwt: [] components: securitySchemes: jwt: type: http scheme: bearer bearerFormat: JWT ````