> ## Documentation Index > Fetch the complete documentation index at: https://koreai.mintlify.site/llms.txt > Use this file to discover all available pages before exploring further. # List session summaries > Use `limit` and `offset` to page through the results. The response includes message, trace-event, and token counts per session, but not the messages or trace details themselves. Use the conversation-history endpoint for messages and the traces endpoint for execution detail. ## OpenAPI ````yaml /agent-platform/apis/sessions.yaml get /api/public/analytics/projects/{projectId}/sessions openapi: 3.1.0 info: title: ABL Public Analytics Sessions API version: 1.0.0 summary: Retrieve project-scoped session analytics. description: > Returns one row per conversation session in a project, summarising how it went: which channel it came in on, how it ended, how many messages were exchanged, how many tokens were used, and what it cost. Give either a date range (`fromDate` and `toDate` together) or a list of `sessionIds`. Other filters are optional. Different filters combine with AND, and multiple values in the same filter combine with OR — so `channel=web_chat,voice&status=ended` means web chat or voice sessions that have ended. Authenticate with the `x-api-key` header. The key must be bound to the project you are querying and have `session:read` permission. x-source-release: release_Artemis_1.5.0 x-source-commit: 267794d75528a9ecb5ed70822451e43f91c15212 servers: - url: https://{host} description: ABL Runtime public endpoint variables: host: default: runtime.example.com security: - ApiKeyAuth: [] tags: - name: Sessions paths: /api/public/analytics/projects/{projectId}/sessions: get: tags: - Sessions summary: List session summaries description: > Use `limit` and `offset` to page through the results. The response includes message, trace-event, and token counts per session, but not the messages or trace details themselves. Use the conversation-history endpoint for messages and the traces endpoint for execution detail. operationId: listPublicAnalyticsSessions parameters: - $ref: '#/components/parameters/ProjectId' - $ref: '#/components/parameters/FromDateOptional' - $ref: '#/components/parameters/ToDateOptional' - $ref: '#/components/parameters/SessionIds' - name: containmentType in: query description: > Filter by how the session was resolved — whether the agent handled it or it went to a human. Pass one or more values, comma-separated or by repeating the parameter, up to 100. Values are case-sensitive and anything else is rejected with `400 INVALID_CONTAINMENT_TYPE`. style: form explode: false schema: type: array maxItems: 100 uniqueItems: true items: type: string enum: - contained - contained_resolved - contained_partial - contained_unresolved - escalated - abandoned - name: environment in: query description: > Filter by environment. Pass one or more values, comma-separated or by repeating the parameter. Values are case-sensitive and anything else is rejected with `400 INVALID_ENVIRONMENT`. Use `dev`, not `development` or `prod` — the service handles both stored spellings for you behind the scenes, but only accepts the four values below as input. style: form explode: false schema: type: array maxItems: 4 uniqueItems: true items: type: string maxLength: 128 enum: - dev - staging - production - working-copy - name: channelUId in: query description: > Filter to sessions belonging to specific end users, using the identity recorded for them on the channel. Pass up to 100 values, comma-separated or by repeating the parameter. style: form explode: false schema: type: array maxItems: 100 uniqueItems: true items: type: string minLength: 1 - name: channel in: query description: > Filter by the channel the conversation came in on. Pass one or more values, comma-separated or by repeating the parameter, up to 50 distinct channels of 128 characters each. This endpoint understands shorthand. Values are grouped into channel families before matching, and each family also matches its older stored spellings, so `channel=web`, `channel=sdk_web`, and `channel=web_chat` all return the same web-chat sessions. The groupings are listed under `x-alias-groups`, and those alias spellings are accepted here even though they are not in the enum — the enum is kept identical across all the public analytics endpoints so there is one channel list to learn. Because of that grouping, asking for one specific voice channel such as `channel=voice_pipeline` returns all voice sessions, not just that transport. Use the traces endpoint if you need to distinguish them. Not restricted to a fixed list, so an unrecognised value returns no results rather than an error. style: form explode: false schema: type: array maxItems: 50 uniqueItems: true items: type: string maxLength: 128 enum: - http_async - slack - line - msteams - whatsapp - messenger - instagram - twilio_sms - zendesk - telegram - genesys - genesys_open_messaging - ai4w - kore_agent_assist - email - voice_vxml - korevg - audiocodes - genesys_audio_connector - voice_pipeline - voice_realtime - voice - voice_twilio - voice_livekit - ag_ui - a2a - sdk_websocket - web_debug - web_chat - api - http x-alias-groups: web_chat: - web - sdk_web - sdk_chat - chat - webchat - sdk_websocket - sdk_ws - sdk-ws - ws web_debug: - debug_websocket - web_debug_socket api: - http - sdk_http - sdk_api twilio_sms: - sms - twilio voice: - vxml - voice_vxml - korevg - audiocodes - genesys_audio_connector - voice_pipeline - voice_realtime - voice_twilio - voice_livekit - twilio_voice - $ref: '#/components/parameters/TraceDimension' - name: status in: query description: > Filter by the session's current state. Pass a single value; if you repeat the parameter only the first value is used. Case-sensitive, and anything else is rejected with `400 INVALID_STATUS`. Note that sessions can be stored with a status of `failed`, but `failed` is not accepted as a filter value here. schema: type: string enum: - active - idle - ended - completed - escalated - abandoned - archived - name: limit in: query description: > How many sessions to return. A page may come back with fewer than requested so the response stays within the 1 MiB size limit — check `hasMore` rather than assuming a short page is the last one. schema: type: integer minimum: 1 maximum: 10000 default: 100 - name: offset in: query schema: type: integer minimum: 0 default: 0 responses: '200': description: Session page returned successfully. content: application/json: schema: $ref: '#/components/schemas/SessionListResponse' example: success: true total: 1 offset: 0 limit: 100 hasMore: false sessions: - id: sess-01J4S8Y1 channel: web_chat environment: production status: ended containmentType: contained_resolved createdAt: '2026-08-01T00:00:00.000Z' lastActivityAt: '2026-08-01T00:05:00.000Z' endedAt: '2026-08-01T00:05:30.000Z' messageCount: user: 3 agent: 4 total: 7 traceEventCount: llm_call: 2 decision: 1 tool_call: 5 error: 1 total: 12 tokenCount: inputTokens: 100 cachedTokens: 10 completionTokens: 32 reasoningTokens: 8 outputTokens: 40 totalTokens: 140 traceDimensions: accountTier: gold estimatedCost: 0.0025 durationMs: 330000 idleDurationMs: 30000 isTest: null '400': $ref: '#/components/responses/BadRequest' '401': $ref: '#/components/responses/Unauthorized' '403': $ref: '#/components/responses/Forbidden' '404': $ref: '#/components/responses/NotFound' '413': $ref: '#/components/responses/ResponseTooLarge' '429': $ref: '#/components/responses/RateLimited' '500': $ref: '#/components/responses/InternalError' '503': $ref: '#/components/responses/Unavailable' components: parameters: ProjectId: name: projectId in: path required: true description: Project identifier bound to the API key. schema: type: string minLength: 1 FromDateOptional: name: fromDate in: query description: >- Inclusive ISO 8601 start timestamp. Required with `toDate` unless `sessionIds` is supplied. schema: type: string format: date-time ToDateOptional: name: toDate in: query description: >- Exclusive ISO 8601 end timestamp. Required with `fromDate` unless `sessionIds` is supplied. schema: type: string format: date-time SessionIds: name: sessionIds in: query description: > Filter to specific sessions. Pass up to 10,000 session IDs, comma-separated or by repeating the parameter. Can be used instead of a date range. style: form explode: false schema: type: array maxItems: 10000 uniqueItems: true items: type: string minLength: 1 TraceDimension: name: traceDimensions[key] in: query description: > Filter by a custom dimension recorded against the session. Replace `key` with the dimension name — for example `traceDimensions[accountTier]=gold`. Repeat with different keys to require all of them. Values match exactly. Remember to URL-encode the brackets if your HTTP client does not do it for you. allowReserved: true schema: type: string x-dynamic-parameter: true schemas: SessionListResponse: type: object additionalProperties: false required: - success - sessions - total - offset - limit - hasMore properties: success: type: boolean const: true description: | Always true on a successful response. sessions: description: The sessions on this page. type: array items: $ref: '#/components/schemas/Session' total: type: integer minimum: 0 description: | Total number of sessions matching your filters, across all pages. offset: type: integer minimum: 0 description: | The offset applied to this response. limit: type: integer minimum: 1 maximum: 10000 description: | The page size applied to this response. hasMore: type: boolean description: | Whether more pages are available after this one. Session: type: object additionalProperties: false required: - id - channel - environment - status - containmentType - createdAt - lastActivityAt - endedAt - messageCount - traceEventCount - tokenCount - traceDimensions - estimatedCost - durationMs - idleDurationMs - isTest properties: id: type: string description: > The session identifier. Use it with the conversation-history and traces endpoints. channel: type: string description: > The channel the conversation came in on, exactly as recorded — not grouped into a family the way the `channel` filter is. Empty when no channel was recorded. enum: - '' - http_async - slack - line - msteams - whatsapp - messenger - instagram - twilio_sms - zendesk - telegram - genesys - genesys_open_messaging - ai4w - kore_agent_assist - email - voice_vxml - korevg - audiocodes - genesys_audio_connector - voice_pipeline - voice_realtime - voice - voice_twilio - voice_livekit - ag_ui - a2a - sdk_websocket - web_debug - web_chat - api - http environment: type: string description: Empty when the source column is null. enum: - '' - development - staging - production - working-copy status: type: string description: > The session's state. `failed` can appear here even though it cannot be used as a `status` filter value. Empty when no status was recorded. enum: - '' - active - idle - ended - completed - failed - escalated - abandoned - archived containmentType: description: > How the session was resolved — whether the agent handled it or it went to a human. type: string enum: - contained - contained_resolved - contained_partial - contained_unresolved - escalated - abandoned createdAt: type: - string - 'null' format: date-time description: | When the session started. lastActivityAt: type: - string - 'null' format: date-time description: | When the last message or event occurred in the session. endedAt: type: - string - 'null' format: date-time description: | When the session ended. Null while it is still open. messageCount: $ref: '#/components/schemas/MessageCount' traceEventCount: $ref: '#/components/schemas/TraceEventCount' tokenCount: $ref: '#/components/schemas/TokenCount' traceDimensions: description: | Custom dimensions recorded against the session, as key/value pairs. Filter on these with the traceDimensions[key] parameter. type: object additionalProperties: type: string estimatedCost: type: number minimum: 0 description: Estimated cost in the platform billing currency. durationMs: type: number minimum: 0 description: | Total elapsed time from start to end, in milliseconds. idleDurationMs: type: number minimum: 0 description: > Part of the duration where nothing happened — time spent waiting, in milliseconds. Useful for separating real handling time from waiting. isTest: type: - boolean - 'null' description: Reserved field; currently returned as null. ErrorEnvelope: type: object required: - error properties: success: type: boolean const: false description: | Always false on an error response. error: description: > Details of what went wrong. `code` is a stable machine-readable value; `message` is human-readable. type: object required: - code - message properties: code: type: string message: type: string details: type: object additionalProperties: true message: type: string description: >- Present on some authorization denials in addition to `error.message`. required: description: >- Permission(s) the caller was missing. Present on authorization denials. oneOf: - type: string - type: array items: type: string MessageCount: type: object additionalProperties: false required: - user - agent - total properties: user: type: number minimum: 0 description: | Messages sent by the end user. agent: type: number minimum: 0 description: | Messages sent by the agent. total: type: number minimum: 0 description: | Total number of trace events recorded for the session. TraceEventCount: type: object additionalProperties: false required: - llm_call - decision - tool_call - error - total properties: llm_call: type: number minimum: 0 description: | Number of LLM calls made during the session. decision: type: number minimum: 0 description: | Number of agent decisions taken, such as routing choices. tool_call: type: number minimum: 0 description: | Number of tool calls made. error: type: number minimum: 0 description: | Number of failed events during the session. total: type: number minimum: 0 description: | Total number of trace events recorded for the session. TokenCount: type: object additionalProperties: false required: - inputTokens - cachedTokens - completionTokens - reasoningTokens - outputTokens - totalTokens properties: inputTokens: type: number minimum: 0 description: | Tokens in the prompts sent to models across the session. cachedTokens: type: number minimum: 0 description: > Input tokens served from the provider prompt cache. Part of inputTokens, and usually cheaper. completionTokens: type: number minimum: 0 description: | Tokens generated as answers, not counting reasoning tokens. reasoningTokens: type: number minimum: 0 description: > Tokens spent on internal model reasoning. Billed as output but not part of any visible answer. outputTokens: type: number minimum: 0 description: | All tokens produced by models — answers plus reasoning. totalTokens: type: number minimum: 0 description: | Input plus output tokens for the whole session. responses: BadRequest: description: >- Invalid or unsupported filters, pagination, date range, or required-filter combination. content: application/json: schema: $ref: '#/components/schemas/ErrorEnvelope' Unauthorized: description: Missing or invalid `x-api-key`, or an Authorization header was supplied. content: application/json: schema: $ref: '#/components/schemas/ErrorEnvelope' Forbidden: description: The key lacks `session:read` for its bound project. content: application/json: schema: $ref: '#/components/schemas/ErrorEnvelope' NotFound: description: >- Project not found or deliberately concealed because it is outside the key's project scope. content: application/json: schema: $ref: '#/components/schemas/ErrorEnvelope' ResponseTooLarge: description: A single session cannot fit within the 1 MiB response budget. content: application/json: schema: $ref: '#/components/schemas/ErrorEnvelope' RateLimited: description: Tenant request rate limit exceeded. content: application/json: schema: $ref: '#/components/schemas/ErrorEnvelope' InternalError: description: Unexpected query failure. content: application/json: schema: $ref: '#/components/schemas/ErrorEnvelope' Unavailable: description: Analytics backing store unavailable. content: application/json: schema: $ref: '#/components/schemas/ErrorEnvelope' securitySchemes: ApiKeyAuth: type: apiKey in: header name: x-api-key description: >- Project-bound API key. Do not send an Authorization header with this API. ````