> ## Documentation Index > Fetch the complete documentation index at: https://www.adaline.ai/docs/llms.txt > Use this file to discover all available pages before exploring further. # Search Spans > Returns a cursor-paginated page of filtered span rows. When `promptId` is provided in the request body, results are scoped to spans that were ingested with that `promptId` — regardless of content type. In practice this is most often `Model` / `ModelStream` spans, but any span type the caller attached to the prompt (for example, a `Tool` span emitted during that prompt's execution) is returned. When `promptId` is omitted, spans of every content type (Model, Tool, Function, Guardrail, Retrieval, Embeddings, Other, or caller-defined custom types) across the project are returned. Pass `nextCursor` from the response as `cursor` to fetch the next page. Returns a cursor-paginated page of filtered model span rows for a prompt. Pass `nextCursor` from the response as `cursor` to fetch the next page. See [Export Logs](/monitor/filter-and-search-logs) for filter types, available columns, and pagination conventions. ## OpenAPI ````yaml POST /logs/spans openapi: 3.1.1 info: title: Adaline API version: 2.0.0 description: >- Public API for managing Adaline projects, prompts, datasets, evaluators, evaluations, providers, models, and deployments. Generated from Zod schemas — this spec is the single source of truth. servers: - url: https://api.adaline.ai/v2 description: Production API security: [] tags: - name: datasets description: Manage prompt datasets and their columns and rows. - name: projects description: List and manage Adaline projects. - name: providers description: List configured LLM providers and their settings. - name: models description: List available models across providers. - name: prompts description: Create and manage prompts and their drafts. - name: evaluators description: Configure evaluators attached to a prompt. - name: evaluations description: Run and inspect evaluation results. - name: deployments description: Retrieve deployed prompt snapshots. - name: logs description: Ingest execution traces and spans for observability. paths: /logs/spans: post: tags: - logs summary: Search spans description: >- Returns a cursor-paginated page of filtered span rows. When `promptId` is provided in the request body, results are scoped to spans that were ingested with that `promptId` — regardless of content type. In practice this is most often `Model` / `ModelStream` spans, but any span type the caller attached to the prompt (for example, a `Tool` span emitted during that prompt's execution) is returned. When `promptId` is omitted, spans of every content type (Model, Tool, Function, Guardrail, Retrieval, Embeddings, Other, or caller-defined custom types) across the project are returned. Pass `nextCursor` from the response as `cursor` to fetch the next page. operationId: searchLogSpans requestBody: required: true content: application/json: schema: $ref: '#/components/schemas/SearchLogSpansRequest' responses: '200': description: A page of search results with cursor-based pagination metadata. content: application/json: schema: $ref: '#/components/schemas/SearchSpansResponse' headers: X-Request-Id: description: >- Unique identifier for the request. Returned in every response. Send a custom value to correlate requests, or omit to receive a generated UUID. schema: type: string format: uuid '400': description: Bad request content: application/json: schema: $ref: '#/components/schemas/ErrorResponse' '401': description: Unauthorized — invalid or missing API key. content: application/json: schema: $ref: '#/components/schemas/ErrorResponse' '403': description: Forbidden — invalid or unauthorized API key. content: application/json: schema: $ref: '#/components/schemas/ErrorResponse' '404': description: Not found — project or prompt not found. content: application/json: schema: $ref: '#/components/schemas/ErrorResponse' '500': description: Internal server error content: application/json: schema: $ref: '#/components/schemas/ErrorResponse' security: - bearerAuth: [] components: schemas: SearchLogSpansRequest: title: Search Log Spans Request type: object description: >- Request payload for paginated span search. `promptId` is optional; when omitted the response includes spans of every content type across the project. required: - projectId properties: projectId: $ref: '#/components/schemas/BaseEntityId' description: The project to search spans in. promptId: $ref: '#/components/schemas/BaseEntityId' description: >- Optional. When provided, scopes results to this prompt. Omit to return spans of every content type (Model, Tool, Function, Guardrail, Retrieval, Embeddings, Other, or custom) across all prompts in the project. model: type: string description: Case-insensitive substring match on the model name. provider: type: string description: Case-insensitive substring match on the provider name. startedAfter: type: integer description: >- Unix timestamp in milliseconds. Only return rows started at or after this time. startedBefore: type: integer description: >- Unix timestamp in milliseconds. Only return rows started at or before this time. status: type: string enum: - success - failure - aborted - cancelled - pending - unknown description: Filter by status. name: type: string description: Case-insensitive substring match on the trace or span name. referenceId: type: string description: Exact match on the reference ID. sessionId: type: string description: Exact match on the session ID. filters: type: array maxItems: 20 items: $ref: '#/components/schemas/LogFilterObject' description: Typed filter objects for advanced queries. ANDed with flat params. sort: type: string enum: - startedAt:asc - startedAt:desc description: Sort order. Defaults to startedAt:desc. limit: type: integer minimum: 1 maximum: 200 description: Page size. Defaults to 50, max 200. cursor: type: string description: Opaque pagination cursor from a previous response. SearchSpansResponse: type: object properties: pagination: $ref: '#/components/schemas/Pagination' spans: type: array items: $ref: '#/components/schemas/LogSpanItem' description: Array of log span items. required: - pagination - spans title: Search Spans Response description: Paginated search results for log spans. ErrorResponse: title: Error Response type: object description: >- Standard error response envelope. Permits backend-specific sibling fields (e.g. Zod issue details) alongside the typed `error`. properties: error: description: >- Error details — typed envelope, or a raw string reason when the backend's validator produces a short summary (e.g. Zod validation fallback). oneOf: - type: string - type: object properties: code: type: string description: Machine-readable error code. enum: - VALIDATION_ERROR - INVALID_CURSOR - AUTHENTICATION_REQUIRED - FORBIDDEN - RESOURCE_NOT_FOUND - CONFLICT - PAYLOAD_TOO_LARGE - RATE_LIMITED - INTERNAL_ERROR message: type: string description: Human-readable error message. details: type: array description: Optional field-level validation details. items: type: object properties: field: type: string issue: type: string required: - field - issue required: - code - message additionalProperties: true required: - error additionalProperties: true BaseEntityId: title: Base Entity ID type: string description: Unique identifier for an entity, between 20 and 80 characters. minLength: 20 maxLength: 80 LogFilterObject: title: Log Filter Object type: object description: >- Typed filter object. Discriminated union on `type`: string, number, datetime, or arrayContains. required: - type - column - operator - value properties: type: type: string enum: - string - number - datetime - arrayContains description: Filter type. Determines valid operators and value types. column: type: string description: Column to filter on. Unknown columns are silently ignored. operator: type: string description: >- Comparison operator. Valid values depend on type: string (eq, contains), number (eq, gt, gte, lt, lte), datetime (gt, gte, lt, lte), arrayContains (contains). value: description: >- Filter value. String for string/arrayContains types, number for number/datetime types. Pagination: title: Pagination type: object description: Pagination metadata for list endpoints with cursor-based navigation. properties: limit: type: integer exclusiveMinimum: 0 returned: type: integer minimum: 0 hasMore: type: boolean nextCursor: type: - string - 'null' required: - limit - returned - hasMore - nextCursor LogSpanItem: title: Log Span Item type: object description: A log span summary returned in search responses. properties: id: type: string description: Unique span identifier. name: type: string description: Name of the span. contentType: type: string description: >- Content type of the span. One of Model, ModelStream, Tool, Function, Guardrail, Retrieval, Embeddings, Other, or a caller-defined custom string. status: type: string description: Span status. enum: - success - failure - aborted - cancelled - unknown startedAt: $ref: '#/components/schemas/UnixTimestamp' description: When the span started. endedAt: $ref: '#/components/schemas/UnixTimestamp' description: When the span ended. latency: type: number description: Span duration in milliseconds. logTraceId: type: string description: ID of the parent trace. promptId: type: - string - 'null' description: >- ID of the associated prompt. Null for non-Model span types (Tool, Function, Guardrail, Retrieval, Embeddings, Other, or custom) that are not tied to a prompt execution. referenceId: type: - string - 'null' description: External reference ID. parentReferenceId: type: - string - 'null' description: Reference ID of the parent span. sessionId: type: - string - 'null' description: Session ID for grouping related spans. cost: type: - number - 'null' description: Cost of this span. promptTokens: type: - number - 'null' description: Input token count. completionTokens: type: - number - 'null' description: Output token count. totalTokens: type: - number - 'null' description: Total token count. content: type: string description: JSON-stringified span content. parsedContent: type: string description: JSON-stringified parsed content. attributes: type: string description: JSON-stringified attributes. tags: type: string description: JSON-stringified tags. events: type: string description: JSON-stringified events. createdAt: $ref: '#/components/schemas/UnixTimestamp' description: When the span was created. updatedAt: $ref: '#/components/schemas/UnixTimestamp' description: When the span was last updated. required: - id - name - contentType - status - startedAt - endedAt - latency - logTraceId - promptId - referenceId - parentReferenceId - sessionId - cost - promptTokens - completionTokens - totalTokens - content - parsedContent - attributes - tags - events - createdAt - updatedAt UnixTimestamp: title: Unix Timestamp type: integer format: int64 description: Unix timestamp in milliseconds. minimum: 1672511400000 maximum: 33229420200000 securitySchemes: bearerAuth: type: http scheme: bearer bearerFormat: API Key description: 'Workspace API key. Pass as `Authorization: Bearer `.' ````