> ## Documentation Index > Fetch the complete documentation index at: https://docs.nimbleway.com/llms.txt > Use this file to discover all available pages before exploring further. # Get Agent Run Result > Fetch the output of a completed run. The `output` is `type: "text"` (a prose answer) or `type: "json"` (structured data matching the output schema), plus `trust` metadata with per-claim citations for the answer. While the run is still `queued` or `running` this endpoint returns `409`; if the run `failed` or was `cancelled` it returns `422` with the run and error details. ## OpenAPI ````yaml /api-reference/openapi.json get /v2/agents/{agent_id}/runs/{run_id}/result openapi: 3.1.0 info: title: Nimble SDK version: 1.0.0 description: The AI-Native SDK for Real-Time Web Data at scale servers: - url: https://sdk.nimbleway.com security: [] paths: /v2/agents/{agent_id}/runs/{run_id}/result: get: tags: - Public API summary: Get Agent Run Result description: >- Fetch the output of a completed run. The `output` is `type: "text"` (a prose answer) or `type: "json"` (structured data matching the output schema), plus `trust` metadata with per-claim citations for the answer. While the run is still `queued` or `running` this endpoint returns `409`; if the run `failed` or was `cancelled` it returns `422` with the run and error details. operationId: >- get_run_result_public_v2_api_v2_public_web_search_agents__agent_id__runs__run_id__result_get parameters: - in: path name: agent_id required: true schema: format: uuid title: Agent Id type: string description: Web Search Agent ID, format `wsa_`. - in: path name: run_id required: true schema: title: Run Id type: string description: Run ID, format `task_run_`. responses: '200': description: Run completed; the output is returned. content: application/json: schema: $ref: '#/components/schemas/TaskRunResultPublicV2' '404': description: Run not found for this agent. content: application/json: schema: type: object properties: detail: type: string title: Detail title: ErrorDetail example: detail: >- Run 'task_run_c0ffee00-0000-4000-8000-000000000000' not found for agent wsa_2af5b0e2-58b8-4b60-a1a4-9e57e4d33f00 '409': description: >- Run is still queued or running. Poll `GET .../runs/{run_id}` until `status` is `completed`, then retry. content: application/json: schema: type: object properties: detail: type: string title: Detail title: ErrorDetail example: detail: >- Run still active; poll the run status endpoint (GET .../runs/task_run_c0ffee00-0000-4000-8000-000000000000) until completed '422': description: >- The run failed or was cancelled. `detail` carries the run and its error. content: application/json: schema: type: object properties: detail: $ref: '#/components/schemas/TaskRunFailedResultPublicV2' title: FailedRunDetail security: - BearerAuth: [] components: schemas: TaskRunResultPublicV2: properties: output: anyOf: - $ref: '#/components/schemas/TaskRunTextOutputPublicV2' - $ref: '#/components/schemas/TaskRunJsonOutputPublicV2' description: Output from the completed task. title: Output run: $ref: '#/components/schemas/TaskRunResponsePublicV2' description: Task run object with status 'completed'. required: - run - output title: TaskRunResultPublicV2 type: object TaskRunFailedResultPublicV2: properties: error: $ref: '#/components/schemas/TaskRunErrorPublicV2' description: Structured error detail. run: $ref: '#/components/schemas/TaskRunResponsePublicV2' description: Task run object with status 'failed'. required: - run - error title: TaskRunFailedResultPublicV2 type: object TaskRunTextOutputPublicV2: properties: content: description: The final prose answer. title: Content type: string trust: $ref: '#/components/schemas/TextRunTrust' description: Trust and citation metadata for the output. type: const: text default: text description: Output content type. title: Type type: string required: - content - trust title: TaskRunTextOutputPublicV2 type: object TaskRunJsonOutputPublicV2: properties: content: anyOf: - additionalProperties: true type: object - items: {} type: array description: The final structured output. title: Content trust: $ref: '#/components/schemas/JsonRunTrust' description: Trust and citation metadata for the output. type: const: json default: json description: Output content type. title: Type type: string required: - content - trust title: TaskRunJsonOutputPublicV2 type: object TaskRunResponsePublicV2: properties: completed_at: anyOf: - format: date-time type: string - type: 'null' description: When the run completed. title: Completed At created_at: description: When the run was created. format: date-time title: Created At type: string effort: $ref: '#/components/schemas/Effort' description: Effort level used for the run. example: high error: anyOf: - $ref: '#/components/schemas/TaskRunErrorPublicV2' - type: 'null' description: Error details when the run failed. id: description: Run identifier, format "task_run_{uuid}". title: Id type: string example: task_run_c0ffee00-0000-4000-8000-000000000000 interaction_id: description: Interaction ID. title: Interaction Id type: string is_active: description: True while status is 'queued' or 'running'. title: Is Active type: boolean prompt: anyOf: - type: string - type: 'null' description: Prompt submitted for the run. title: Prompt started_at: anyOf: - format: date-time type: string - type: 'null' description: When the run started executing. title: Started At status: $ref: '#/components/schemas/TaskRunStatusValue' description: Current run status. example: queued web_search_agent_id: description: Web Search Agent instance this run belongs to. title: Web Search Agent Id type: string example: wsa_2af5b0e2-58b8-4b60-a1a4-9e57e4d33f00 required: - id - interaction_id - status - is_active - effort - created_at - web_search_agent_id title: TaskRunResponsePublicV2 type: object TaskRunErrorPublicV2: properties: message: description: Human-readable error description. title: Message type: string ref_id: description: Reference ID (equals the run id). title: Ref Id type: string required: - ref_id - message title: TaskRunErrorPublicV2 type: object TextRunTrust: description: >- Trust metadata for a prose answer: run-level confidence plus per-claim citations. properties: claims: description: Per-claim trust, keyed by callout markers in the answer text. items: $ref: '#/components/schemas/TextClaimTrust' title: Claims type: array confidence: $ref: '#/components/schemas/TrustConfidence' description: Overall confidence in the answer. reasoning: description: Why this confidence level was assigned. title: Reasoning type: string sources: description: Sources consulted while producing the answer. items: $ref: '#/components/schemas/TrustSource' title: Sources type: array required: - sources - confidence - reasoning - claims title: TextRunTrust type: object JsonRunTrust: description: >- Trust metadata for a structured answer: run-level confidence plus per-value citations. properties: claims: description: Per-value trust, keyed by JSON path in the structured output. items: $ref: '#/components/schemas/JsonClaimTrust' title: Claims type: array confidence: $ref: '#/components/schemas/TrustConfidence' description: Overall confidence in the answer. reasoning: description: Why this confidence level was assigned. title: Reasoning type: string sources: description: Sources consulted while producing the answer. items: $ref: '#/components/schemas/TrustSource' title: Sources type: array required: - sources - confidence - reasoning - claims title: JsonRunTrust type: object Effort: description: Canonical effort tier names for the research graph. enum: - low - medium - high - x-high - max title: Effort type: string TaskRunStatusValue: description: >- Lowercase status values used in API responses (distinct from the DB-level TaskRunStatus enum). enum: - queued - running - completed - failed - cancelled title: TaskRunStatusValue type: string TextClaimTrust: description: Trust metadata for one claim in a prose answer, keyed by callout marker. properties: callout: description: Callout marker number referencing this claim in the answer text. title: Callout type: integer citations: description: Citations backing this claim. items: $ref: '#/components/schemas/TrustCitation' title: Citations type: array confidence: $ref: '#/components/schemas/TrustConfidence' description: Confidence in this claim. reasoning: description: Why this confidence level was assigned. title: Reasoning type: string required: - citations - confidence - reasoning - callout title: TextClaimTrust type: object TrustConfidence: enum: - high - medium - low - pre_existing title: TrustConfidence type: string TrustSource: description: A source consulted while producing the answer. properties: extract_template_name: anyOf: - type: string - type: 'null' description: Extract template used to read the source, when one was used. title: Extract Template Name source_category: anyOf: - $ref: '#/components/schemas/SourceCategory' - type: 'null' description: >- Kind of source: official, news, social, academic, aggregator, or other. source_intent: anyOf: - $ref: '#/components/schemas/SourceCategory' - type: 'null' description: Kind of source the underlying question asked for. title: anyOf: - type: string - type: 'null' description: Title of the source page. title: Title type: $ref: '#/components/schemas/TrustSourceType' description: 'How authoritative the source is: ''primary'' or ''secondary''.' url: description: URL of the source page. title: Url type: string required: - url - type title: TrustSource type: object JsonClaimTrust: description: >- Trust metadata for one value in a structured (JSON) answer, keyed by JSON path. properties: citations: description: Citations backing this value. items: $ref: '#/components/schemas/TrustCitation' title: Citations type: array confidence: $ref: '#/components/schemas/TrustConfidence' description: Confidence in this value. path: description: >- JSON path of the value in the structured output this claim refers to. title: Path type: string reasoning: description: Why this confidence level was assigned. title: Reasoning type: string required: - citations - confidence - reasoning - path title: JsonClaimTrust type: object TrustCitation: description: A citation backing a specific claim in the answer. properties: excerpts: anyOf: - items: type: string type: array - type: 'null' description: Verbatim excerpts supporting the claim. title: Excerpts example: - Acme raised a $30M Series B in June 2026. extract_template_name: anyOf: - type: string - type: 'null' description: Extract template used to read the source, when one was used. title: Extract Template Name source_category: anyOf: - $ref: '#/components/schemas/SourceCategory' - type: 'null' description: >- Kind of source: official, news, social, academic, aggregator, or other. source_intent: anyOf: - $ref: '#/components/schemas/SourceCategory' - type: 'null' description: Kind of source the underlying question asked for. source_type: anyOf: - $ref: '#/components/schemas/TrustSourceType' - type: 'null' description: 'How authoritative the source is: ''primary'' or ''secondary''.' title: anyOf: - type: string - type: 'null' description: Title of the cited page. title: Title url: description: URL of the cited page. title: Url type: string example: https://www.linkedin.com/company/acme required: - url title: TrustCitation type: object SourceCategory: description: >- What *kind* of source this is (classified by the compress LLM), independent of TrustSourceType (how authoritative it is for a specific claim). Deliberately uses "official" rather than "primary" so the two axes can never collide. Also doubles as the sub-question's ``source_intent`` (what kind of source a question *needs*) — the two concepts overlap enough that a single enum lets ``classify_source_importance`` compare "what we got" against "what we asked for" directly. enum: - official - news - social - academic - aggregator - other title: SourceCategory type: string TrustSourceType: enum: - primary - secondary title: TrustSourceType type: string securitySchemes: BearerAuth: type: http scheme: bearer ````