> ## 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 LLM calls > `fromDate` and `toDate` are required. `fromDate` is included in the results and `toDate` is not, so a full day is `fromDate=2026-08-04T00:00:00Z&toDate=2026-08-05T00:00:00Z`. All other filters are optional and match exactly. Filters combine with AND, so adding more narrows the results. `source` and `sourceSubtype` also accept older spellings and translate them for you — see those parameters for the details. ## OpenAPI ````yaml /agent-platform/apis/llm-ledger.yaml get /api/public/analytics/projects/{projectId}/llm-ledger openapi: 3.1.0 info: title: ABL Public Analytics LLM Ledger API version: 1.0.0 summary: Retrieve a project-scoped ledger of individual LLM calls. description: > Returns one row for every individual LLM call made in a project, with its model, provider, token counts, latency, and cost. Calls are listed individually — nothing is rolled up per conversation turn. Use `dataMode` to control detail. `summary` returns the facts about each call. `full` adds the exact prompt sent and response received under each call's `payload`, so you do not need a second request to retrieve them. Authenticate with the `x-api-key` header. The key must be bound to the project you are querying and have `analytics:read` permission. x-source-release: release_Artemis_1.5.0 x-source-commit: 267794d75528a9ecb5ed70822451e43f91c15212 servers: - url: https://{host} variables: host: default: runtime.example.com security: - ApiKeyAuth: [] tags: - name: LLM Ledger paths: /api/public/analytics/projects/{projectId}/llm-ledger: get: tags: - LLM Ledger summary: List LLM calls description: > `fromDate` and `toDate` are required. `fromDate` is included in the results and `toDate` is not, so a full day is `fromDate=2026-08-04T00:00:00Z&toDate=2026-08-05T00:00:00Z`. All other filters are optional and match exactly. Filters combine with AND, so adding more narrows the results. `source` and `sourceSubtype` also accept older spellings and translate them for you — see those parameters for the details. operationId: listPublicLlmLedger parameters: - $ref: '#/components/parameters/ProjectId' - name: fromDate in: query required: true schema: type: string format: date-time - name: toDate in: query required: true schema: type: string format: date-time - name: limit in: query description: > How many calls to return. When `dataMode=full`, a page may come back with fewer rows 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 - name: sessionIds in: query description: > Filter to specific sessions. Pass up to 10,000 session IDs, comma-separated or by repeating the parameter. style: form explode: false schema: type: array maxItems: 10000 uniqueItems: true items: type: string minLength: 1 - name: channel in: query description: > Filter by the channel the conversation took place on. Design-time calls, such as Studio test calls, are not tied to a channel and will not match any channel filter. Matched exactly as stored, with no alias handling. Unlike the sessions endpoint, shorthand such as `web` or `sms` is not expanded here — use the exact channel name such as `web_chat` or `twilio_sms`. Not restricted to a fixed list, so an unrecognised value returns no results rather than an error. Cannot be empty (`400 INVALID_FILTER`). schema: type: string 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 - name: source in: query description: > Filter by which part of the platform made the call — for example `runtime` for live agent traffic, or `evals` for evaluation runs. You can pass either the current name or an older one; the older spellings listed under `x-alias-groups` are translated for you, and matching is case-insensitive with `-` and spaces treated as `_`. So `agent_session` finds the same calls as `runtime`. Not restricted to a fixed list, so an unrecognised value returns no results rather than an error. Cannot be empty (`400 INVALID_FILTER`). schema: type: string enum: - runtime - runtime_internal - arch - evals - pipelines - guardrails - knowledge - studio_test_calls - channels - agent_session - eval - analytics_pipeline - analytics_query - guardrail - search_ai - prompt_library - model_health - model_admin - health:credential-store - sdk_widget_localization x-alias-groups: runtime: - runtime - agent_session runtime_internal: - runtime_internal arch: - arch evals: - evals - eval pipelines: - pipelines - analytics_pipeline guardrails: - guardrails - guardrail knowledge: - knowledge - search_ai - analytics_query studio_test_calls: - studio_test_calls - prompt_library - model_health - model_admin - health:credential-store channels: - channels - sdk_widget_localization - name: sourceSubtype in: query description: > Filter by the specific purpose of the call within its source — for example `response_generation` for replies to users, or `guardrails` for safety checks. Older spellings are translated for you, as listed under `x-alias-groups`. If you also pass `source`, translation is limited to that source's subtypes; on its own, subtype names are resolved across all sources. Not restricted to a fixed list, so an unrecognised value returns no results rather than an error. Cannot be empty (`400 INVALID_FILTER`). schema: type: string enum: - response_generation - conversation_history_compaction - ner_gather_llm_extraction - model_based_pii_entity_recognizer - model_entity_preview - llm_backed_validation - searchai_kb_query_intelligence - searchai_query_assistance - routing_pipeline - pipeline_classifier - pipeline_tool_filter - pipeline_merge - contextual_filler_generation - tool_result_summarization - sip_handoff_summary - semantic_execution - legacy_nlu_fallback - general_assistance - project_assistance - eval_remediation - evaluation_suite_generation - natural_language_analytics_query - text_to_sql - guardrails - studio_test_calls - widget_copy_translation - tool_use_iteration - response_gen - realtime_response - conversation_compaction - nlu_entity_extraction - kb_classify_rewrite - field_validation - agent_building - eval_suite_creation - eval_design_time_asset_generation - nl_to_sql - nl_query - query_intelligence - prompt_test - prompt_experiment - credential_health_check - translation_generation x-alias-groups: response_generation: - tool_use_iteration - response_gen - realtime_response conversation_history_compaction: - conversation_compaction ner_gather_llm_extraction: - nlu_entity_extraction searchai_kb_query_intelligence: - kb_classify_rewrite llm_backed_validation: - field_validation general_assistance: - agent_building evaluation_suite_generation: - eval_suite_creation - eval_design_time_asset_generation natural_language_analytics_query: - nl_to_sql - nl_query text_to_sql: - nl_to_sql searchai_query_assistance: - query_intelligence studio_test_calls: - prompt_test - prompt_experiment - credential_health_check widget_copy_translation: - translation_generation - name: status in: query description: > Filter to successful or failed calls. Any other value is rejected with `400 INVALID_STATUS`. schema: type: string enum: - success - failed - name: model in: query description: > Filter by model, using the exact model id as recorded, such as `gpt-4o`. There is no fixed list, because each tenant can register its own models. Calls recorded without a model id appear as `unknown` and can be found with `model=unknown`. Cannot be empty (`400 INVALID_FILTER`). schema: type: string minLength: 1 - name: provider in: query description: > Filter by model provider, such as `openai` or `azure`. Calls recorded without a provider appear as `unknown` and can be found with `provider=unknown`. Not restricted to a fixed list, so an unrecognised value returns no results rather than an error. The values below are the providers the platform can call. Cannot be empty (`400 INVALID_FILTER`). schema: type: string minLength: 1 x-common-values: - openai - anthropic - azure - bedrock - microsoft_foundry - google - gemini - vertex - vertex_ai - google_vertex - cohere - groq - mistral - fireworks - togetherai - together - perplexity - deepseek - xai - openrouter - litellm - openai_compatible - ultravox - custom - mock - unknown - name: environment in: query description: > Filter by environment. Note that this endpoint stores `development` rather than `dev` — unlike the sessions endpoint, which uses `dev`. Design-time calls, such as Studio test calls, are not tied to an environment and will not match this filter. Matched exactly as stored. Not restricted to a fixed list, so an unrecognised value returns no results rather than an error. Cannot be empty (`400 INVALID_FILTER`). schema: type: string enum: - development - staging - production - working-copy - name: name in: query description: Exact effective agent name, operation type, or provider fallback. schema: type: string minLength: 1 - name: channelType in: query description: Exact transport-specific channel type. schema: type: string 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 - name: dataMode in: query description: > How much detail to return per call. `summary` returns the facts about each call. `full` also includes the prompt sent and the response received, under `payload`, where these were stored. schema: type: string enum: - summary - full default: summary - name: sortBy in: query schema: type: string enum: - timestamp - latencyMs - cost - totalTokens default: timestamp - name: sortOrder in: query schema: type: string enum: - asc - desc default: asc responses: '200': description: LLM ledger page returned successfully. headers: Cache-Control: schema: type: string example: private, no-store, max-age=0 content: application/json: schema: $ref: '#/components/schemas/LlmLedgerResponse' examples: summary: summary: Summary mode value: success: true statusCode: 200 total: 1 limit: 100 offset: 0 dataMode: summary hasMore: false generations: - id: >- sess-1:call-1:2026-08-04T09:59:36.316Z:GPT-5.2:azure:1396:780:11659:0 model: GPT-5.2 agentName: Support Agent provider: azure environment: production status: success inputTokens: 1396 cachedTokens: 224 completionTokens: 500 reasoningTokens: 280 outputTokens: 780 totalTokens: 2176 latencyMs: 11659 timeToFirstTokenMs: 0 streamingUsed: false cost: 0.42 timestamp: '2026-08-04T09:59:36.316Z' sessionId: sess-1 callId: call-1 traceId: trace-1 eventId: event-1 source: runtime sourceSubtype: response_generation sourceLabel: Runtime sourceSubtypeLabel: Response Generation sourceId: agent-1 channel: web_chat channelType: web_chat statusCode: 200 payloadState: available full: summary: Full mode with inline request and response value: success: true statusCode: 200 total: 1 limit: 100 offset: 0 dataMode: full hasMore: false generations: - id: >- sess-1:call-1:2026-08-04T09:59:36.316Z:GPT-5.2:azure:1396:780:11659:0 model: GPT-5.2 agentName: Support Agent provider: azure environment: production status: success inputTokens: 1396 cachedTokens: 224 completionTokens: 500 reasoningTokens: 280 outputTokens: 780 totalTokens: 2176 latencyMs: 11659 timeToFirstTokenMs: 0 streamingUsed: false cost: 0.42 timestamp: '2026-08-04T09:59:36.316Z' sessionId: sess-1 callId: call-1 traceId: trace-1 eventId: event-1 source: runtime sourceSubtype: response_generation sourceLabel: Runtime sourceSubtypeLabel: Response Generation sourceId: agent-1 channel: web_chat channelType: web_chat statusCode: 200 payloadState: available payload: request: model: GPT-5.2 messages: - role: user content: Hello response: text: How can I help? '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 schema: type: string minLength: 1 schemas: LlmLedgerResponse: type: object additionalProperties: false required: - success - statusCode - total - limit - offset - dataMode - hasMore - generations properties: success: type: boolean const: true description: | Always true on a successful response. statusCode: type: integer const: 200 description: > HTTP status code returned by the provider — 200 when the call succeeded. total: type: integer minimum: 0 description: | Total number of calls matching your filters, across all pages. limit: type: integer minimum: 1 maximum: 10000 description: | The page size applied to this response. offset: type: integer minimum: 0 description: | The offset applied to this response. dataMode: type: string enum: - summary - full description: | The detail level applied to this response, echoing your request. hasMore: type: boolean description: | Whether more pages are available after this one. generations: type: array items: $ref: '#/components/schemas/LlmGeneration' description: | The LLM calls on this page. LlmGeneration: type: object additionalProperties: false required: - id - model - agentName - provider - environment - status - inputTokens - cachedTokens - completionTokens - reasoningTokens - outputTokens - totalTokens - latencyMs - timeToFirstTokenMs - streamingUsed - cost - timestamp - sessionId - callId - traceId - eventId - source - sourceSubtype - sourceLabel - sourceSubtypeLabel - sourceId - channel - channelType - statusCode - payloadState properties: id: type: string description: Stable composite identifier derived from call attributes. model: type: string description: > The model used, as recorded — for example `gpt-4o`. There is no fixed list, since each tenant can register its own models. `unknown` when no model was recorded. minLength: 1 agentName: type: string description: Agent name falling back to operation type or provider.: null provider: type: string description: > The model provider, such as `openai` or `azure`. `unknown` when no provider was recorded. The values below are the providers the platform can call, but this is not a fixed list. minLength: 1 x-common-values: - openai - anthropic - azure - bedrock - microsoft_foundry - google - gemini - vertex - vertex_ai - google_vertex - cohere - groq - mistral - fireworks - togetherai - together - perplexity - deepseek - xai - openrouter - litellm - openai_compatible - ultravox - custom - mock - unknown environment: type: string description: > The environment the call ran in. Empty for design-time calls that have no conversation behind them, such as Arch, evaluation runs, pipelines, and Studio test calls. enum: - '' - development - staging - production - working-copy status: type: string enum: - success - failed description: | Whether the call succeeded or failed. inputTokens: type: number minimum: 0 description: | Tokens in the prompt sent to the model. cachedTokens: type: number minimum: 0 description: > Input tokens served from the provider's prompt cache. These are part of inputTokens and usually cost less. completionTokens: type: number minimum: 0 description: > Tokens the model generated as its answer, not counting reasoning tokens. reasoningTokens: type: number minimum: 0 description: > Tokens the model spent on internal reasoning. Billed as output but not part of the visible answer. outputTokens: type: number minimum: 0 description: | All tokens the model produced — the answer plus any reasoning. totalTokens: type: number minimum: 0 description: | Input plus output tokens for this call. latencyMs: type: number minimum: 0 description: | Total time the call took, in milliseconds. timeToFirstTokenMs: type: - number - 'null' minimum: 0 description: > Time until the first token arrived, in milliseconds. Meaningful for streamed calls; null otherwise. streamingUsed: type: boolean description: | Whether the response was streamed back token by token. cost: type: number minimum: 0 description: >- Estimated cost. No currency field exists on this endpoint; values are implicitly USD. timestamp: type: string format: date-time description: | When the call was made. sessionId: type: string description: > The conversation this call belongs to. Empty for calls made outside a conversation, such as Studio test calls. callId: type: string description: | Identifier of the individual LLM call. traceId: type: string description: > The trace this call belongs to. Use it to look the call up in the traces API. eventId: type: string description: > The trace event that recorded this call. Use it with the traces API for the full execution context. source: type: string description: > Which part of the platform made the call. Older recorded spellings are translated to the current name where possible; anything unrecognised is returned as recorded. enum: - runtime - runtime_internal - arch - evals - pipelines - guardrails - knowledge - studio_test_calls - channels sourceSubtype: type: string description: > What the call was used for within its source, such as `response_generation`. Older recorded spellings are translated to the current name where possible; anything unrecognised is returned as recorded. enum: - response_generation - conversation_history_compaction - ner_gather_llm_extraction - model_based_pii_entity_recognizer - model_entity_preview - llm_backed_validation - searchai_kb_query_intelligence - searchai_query_assistance - routing_pipeline - pipeline_classifier - pipeline_tool_filter - pipeline_merge - contextual_filler_generation - tool_result_summarization - sip_handoff_summary - semantic_execution - legacy_nlu_fallback - general_assistance - project_assistance - eval_remediation - evaluation_suite_generation - natural_language_analytics_query - text_to_sql - guardrails - studio_test_calls - widget_copy_translation sourceLabel: type: string description: > A display-ready name for `source`, such as `Studio test calls`. If the source is not one of the known values, its raw value is returned here instead. enum: - Runtime - Runtime Internal - Arch - Evals - Pipelines - Guardrails - Knowledge - Studio test calls - Channels sourceSubtypeLabel: type: string description: > A display-ready name for `sourceSubtype`, such as `Response Generation`. The 25 labels below cover the known subtypes; anything else is returned as its raw value, so this list is not exhaustive. Note that `routing_pipeline` currently has no display name and is returned as `routing_pipeline`. x-common-values: - Response Generation - Conversation history compaction - NER / gather LLM extraction - Model-based PII/entity recognizer - Model entity preview - LLM-backed validation - SearchAI KB Query Intelligence - Pipeline classifier - Pipeline tool filter - Pipeline merge - Contextual filler generation - Tool-result summarization - SIP handoff summary - Semantic execution - SearchAI query assistance - Legacy NLU fallback - General assistance - Project assistance - Eval remediation - Evaluation suite generation - Natural-language analytics query - Text-to-SQL - Guardrails - Studio test calls - Widget copy translation x-label-map: response_generation: Response Generation conversation_history_compaction: Conversation history compaction ner_gather_llm_extraction: NER / gather LLM extraction model_based_pii_entity_recognizer: Model-based PII/entity recognizer model_entity_preview: Model entity preview llm_backed_validation: LLM-backed validation searchai_kb_query_intelligence: SearchAI KB Query Intelligence pipeline_classifier: Pipeline classifier pipeline_tool_filter: Pipeline tool filter pipeline_merge: Pipeline merge contextual_filler_generation: Contextual filler generation tool_result_summarization: Tool-result summarization sip_handoff_summary: SIP handoff summary semantic_execution: Semantic execution searchai_query_assistance: SearchAI query assistance legacy_nlu_fallback: Legacy NLU fallback general_assistance: General assistance project_assistance: Project assistance eval_remediation: Eval remediation evaluation_suite_generation: Evaluation suite generation natural_language_analytics_query: Natural-language analytics query text_to_sql: Text-to-SQL guardrails: Guardrails studio_test_calls: Studio test calls widget_copy_translation: Widget copy translation sourceId: type: string description: | Identifier of whatever made the call — usually the agent ID. channel: type: string description: > The channel the conversation took place on, exactly as recorded. Empty for calls with no conversation behind them, such as Studio test calls. 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 channelType: type: string description: Empty for calls made outside a channel session. 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 statusCode: type: integer minimum: 100 maximum: 599 description: > HTTP status code returned by the provider — 200 when the call succeeded. payloadState: type: string enum: - available - omitted - missing - unknown description: > Whether the prompt and response were kept for this call: `available` (returned when dataMode=full), `omitted` (kept but not returned), `missing` (never kept), `unknown` (not recorded). payload: $ref: '#/components/schemas/LlmPayload' description: >- Present only in full mode when at least one stored payload side exists. 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 LlmPayload: type: object additionalProperties: false required: - request - response properties: request: description: Exact stored request payload or null when only a response is stored.: null response: description: Exact stored response payload or null when only a request is stored.: null responses: BadRequest: description: > The request could not be understood. Common causes: `fromDate` or `toDate` missing or not a valid timestamp, `fromDate` not earlier than `toDate`, an unrecognised query parameter, an empty filter value, an invalid value for `status`, `dataMode`, `sortBy` or `sortOrder`, or a `limit` or `offset` outside the allowed range. The `error.code` says which. content: application/json: schema: $ref: '#/components/schemas/ErrorEnvelope' Unauthorized: description: Missing or invalid `x-api-key`, or Authorization was supplied. content: application/json: schema: $ref: '#/components/schemas/ErrorEnvelope' Forbidden: description: The key lacks `analytics:read`. content: application/json: schema: $ref: '#/components/schemas/ErrorEnvelope' NotFound: description: Project not found or concealed because it is outside the key's scope. content: application/json: schema: $ref: '#/components/schemas/ErrorEnvelope' ResponseTooLarge: description: One full-mode generation exceeds 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 LLM ledger query failure. content: application/json: schema: $ref: '#/components/schemas/ErrorEnvelope' Unavailable: description: LLM ledger 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. ````