--- title: List events for an Agent Observability experiment description: Retrieve spans and experiment-level summary metrics for a given experiment with cursor-based pagination. breadcrumbs: Docs > API > List events for an Agent Observability experiment --- > For the complete documentation index, see [llms.txt](https://dnd37xeb23o8n.cloudfront.net/stefon.simmons/docs-astro-web-10435/llms.txt). # List events for an Agent Observability experiment ## v2 (latest) {% alert level="warning" %} **Note**: This endpoint is in preview and is subject to change. If you have any feedback, contact [Datadog support](https://docs.datadoghq.com/help.md). {% /alert %} | Datadog site | API endpoint | | ----------------- | -------------------------------------------------------------------------------------- | | app.datadoghq.com | GET https://api.datadoghq.com/api/v2/llm-obs/v3/experiments/{experiment_id}/events | | us3.datadoghq.com | GET https://api.us3.datadoghq.com/api/v2/llm-obs/v3/experiments/{experiment_id}/events | | us5.datadoghq.com | GET https://api.us5.datadoghq.com/api/v2/llm-obs/v3/experiments/{experiment_id}/events | | app.datadoghq.eu | GET https://api.datadoghq.eu/api/v2/llm-obs/v3/experiments/{experiment_id}/events | | ap1.datadoghq.com | GET https://api.ap1.datadoghq.com/api/v2/llm-obs/v3/experiments/{experiment_id}/events | | ap2.datadoghq.com | GET https://api.ap2.datadoghq.com/api/v2/llm-obs/v3/experiments/{experiment_id}/events | | uk1.datadoghq.com | GET https://api.uk1.datadoghq.com/api/v2/llm-obs/v3/experiments/{experiment_id}/events | | app.ddog-gov.com | GET https://api.ddog-gov.com/api/v2/llm-obs/v3/experiments/{experiment_id}/events | | us2.ddog-gov.com | GET https://api.us2.ddog-gov.com/api/v2/llm-obs/v3/experiments/{experiment_id}/events | ### Overview Retrieve spans and experiment-level summary metrics for a given experiment with cursor-based pagination. ### Arguments #### Path Parameters | Field | Type | Description | | -------------------------- | ------ | --------------------------------------------- | | experiment_id [*required*] | string | The ID of the Agent Observability experiment. | #### Query Strings | Field | Type | Description | | ------------ | ------- | ------------------------------------------------------------------------------ | | page[limit] | integer | Maximum number of spans to return per page. Defaults to 5000. Default: `5000`. | | page[cursor] | string | Opaque cursor from a previous response to fetch the next page of results. | ### Response {% tabs %} {% tab label="200" %} OK {% tabs %} {% tab label="Model" %} Response for listing experiment events (v2/v3). Returns spans and summary metrics in a single resource. | Parent field | Field | Type | Description | | -------------------- | ---------------------------- | -------------- | ------------------------------------------------------------------------------------------------------------------------------ | | | data [*required*] | object | JSON:API data object for an experiment events response. | | data | attributes [*required*] | object | Attributes of an experiment events response. | | attributes | spans [*required*] | [object] | Experiment spans, each enriched with their associated evaluation metrics. | | spans | dataset_record_id | string | ID of the dataset record this span evaluated. | | spans | duration | double | Duration of the span in nanoseconds. | | spans | eval_metrics | [object] | Evaluation metrics associated with this span. | | eval_metrics | assessment | enum | Assessment result for an Agent Observability experiment metric. Allowed enum values: `pass`, `fail`. | | eval_metrics | boolean_value | boolean | Boolean value. Present when `metric_type` is `boolean`. | | eval_metrics | categorical_value | string | Categorical value. Present when `metric_type` is `categorical`. | | eval_metrics | eval_source_type | string | Source type of the evaluation. | | eval_metrics | id | string | Unique identifier of the evaluation metric event. | | eval_metrics | json_value | object | JSON value. Present when `metric_type` is `json`. | | eval_metrics | label | string | Label or name for the metric. | | eval_metrics | metadata | object | Arbitrary metadata associated with the metric. | | eval_metrics | metric_source | string | Source of the metric. Either `custom` (user-submitted) or `summary` (experiment-level aggregate). | | eval_metrics | metric_type | enum | Type of metric recorded for an Agent Observability experiment. Allowed enum values: `score`, `categorical`, `boolean`, `json`. | | eval_metrics | reasoning | string | Human-readable reasoning for the metric value. | | eval_metrics | score_value | double | Numeric score. Present when `metric_type` is `score`. | | eval_metrics | span_id | string | Span ID this metric is associated with. | | eval_metrics | tags | [string] | Tags associated with the metric. | | eval_metrics | timestamp_ms | int64 | Timestamp when the metric was recorded, in milliseconds since Unix epoch. | | eval_metrics | trace_id | string | Trace ID linking this metric to a span. | | spans | id | string | Unique identifier of the span. | | spans | meta | object | Metadata associated with an experiment span. | | meta | error | object | Error details for an experiment span. | | error | message | string | Error message. | | error | stack | string | Stack trace of the error. | | error | type | string | The error type or exception class name. | | meta | expected_output | object | Expected output for the span, used for evaluation. | | meta | input | object | Represents any valid JSON value. | | input | Object 1 | string | A scalar value represented as a string. | | input | Object 2 | double | A scalar numeric value. | | input | Object 3 | object | An arbitrary object value with additional properties. | | input | Object 4 | [] | An array of arbitrary values. | | Object 4 | Object 1 | string | A scalar value represented as a string. | | Object 4 | Object 2 | double | A scalar numeric value. | | Object 4 | Object 3 | object | An arbitrary object value with additional properties. | | Object 4 | Object 4 | boolean | A scalar boolean value. | | input | Object 5 | boolean | A scalar boolean value. | | meta | output | object | Represents any valid JSON value. | | output | Object 1 | string | A scalar value represented as a string. | | output | Object 2 | double | A scalar numeric value. | | output | Object 3 | object | An arbitrary object value with additional properties. | | output | Object 4 | [] | An array of arbitrary values. | | Object 4 | Object 1 | string | A scalar value represented as a string. | | Object 4 | Object 2 | double | A scalar numeric value. | | Object 4 | Object 3 | object | An arbitrary object value with additional properties. | | Object 4 | Object 4 | boolean | A scalar boolean value. | | output | Object 5 | boolean | A scalar boolean value. | | spans | metrics | object | Numeric metrics attached to the span. | | additionalProperties | | double | | | spans | name | string | Name of the span. | | spans | parent_id | string | Parent span ID, if any. | | spans | span_id | string | Span ID. | | spans | start_ns | int64 | Start time in nanoseconds since Unix epoch. | | spans | status | enum | Status of the span. Allowed enum values: `ok`, `error`. | | spans | tags | [string] | Tags associated with the span. | | spans | trace_id | string | Trace ID. | | attributes | summary_metrics [*required*] | [object] | Experiment-level summary evaluation metrics (not tied to individual spans). | | summary_metrics | assessment | enum | Assessment result for an Agent Observability experiment metric. Allowed enum values: `pass`, `fail`. | | summary_metrics | boolean_value | boolean | Boolean value. Present when `metric_type` is `boolean`. | | summary_metrics | categorical_value | string | Categorical value. Present when `metric_type` is `categorical`. | | summary_metrics | eval_source_type | string | Source type of the evaluation. | | summary_metrics | id | string | Unique identifier of the evaluation metric event. | | summary_metrics | json_value | object | JSON value. Present when `metric_type` is `json`. | | summary_metrics | label | string | Label or name for the metric. | | summary_metrics | metadata | object | Arbitrary metadata associated with the metric. | | summary_metrics | metric_source | string | Source of the metric. Either `custom` (user-submitted) or `summary` (experiment-level aggregate). | | summary_metrics | metric_type | enum | Type of metric recorded for an Agent Observability experiment. Allowed enum values: `score`, `categorical`, `boolean`, `json`. | | summary_metrics | reasoning | string | Human-readable reasoning for the metric value. | | summary_metrics | score_value | double | Numeric score. Present when `metric_type` is `score`. | | summary_metrics | span_id | string | Span ID this metric is associated with. | | summary_metrics | tags | [string] | Tags associated with the metric. | | summary_metrics | timestamp_ms | int64 | Timestamp when the metric was recorded, in milliseconds since Unix epoch. | | summary_metrics | trace_id | string | Trace ID linking this metric to a span. | | data | id [*required*] | string | Identifier for this events resource. | | data | type [*required*] | enum | Resource type for an experiment events collection. Allowed enum values: `experiment_events`. | | | meta | object | Pagination cursor metadata. | | meta | after | string | Cursor for the next page of results. | {% /tab %} {% tab label="Example" %} ```json { "data": { "attributes": { "spans": [ { "dataset_record_id": "string", "duration": 1500000000, "eval_metrics": [ { "assessment": "pass", "boolean_value": false, "categorical_value": "string", "eval_source_type": "managed", "id": "00000000-0000-0000-0000-000000000001", "json_value": {}, "label": "faithfulness", "metadata": {}, "metric_source": "custom", "metric_type": "score", "reasoning": "string", "score_value": "number", "span_id": "span-7a1b2c3d", "tags": [], "timestamp_ms": 1705314600000, "trace_id": "abc123def456" } ], "id": "00000000-0000-0000-0000-000000000001", "meta": { "error": { "message": "Model response timed out", "stack": "Traceback (most recent call last):\n File \"main.py\", line 10, in \n response = model.generate(input)\n File \"model.py\", line 45, in generate\n raise TimeoutError(\"Model response timed out\")\nTimeoutError: Model response timed out", "type": "TimeoutError" }, "expected_output": {}, "input": { "description": "undefined", "type": "undefined" }, "output": { "description": "undefined", "type": "undefined" } }, "metrics": { "": "number" }, "name": "llm_call", "parent_id": "string", "span_id": "span-7a1b2c3d", "start_ns": 1705314600000000000, "status": "ok", "tags": [], "trace_id": "abc123def456" } ], "summary_metrics": [ { "assessment": "pass", "boolean_value": false, "categorical_value": "string", "eval_source_type": "managed", "id": "00000000-0000-0000-0000-000000000001", "json_value": {}, "label": "faithfulness", "metadata": {}, "metric_source": "custom", "metric_type": "score", "reasoning": "string", "score_value": "number", "span_id": "span-7a1b2c3d", "tags": [], "timestamp_ms": 1705314600000, "trace_id": "abc123def456" } ] }, "id": "3fd6b5e0-8910-4b1c-a7d0-5b84de329012", "type": "experiment_events" }, "meta": { "after": "string" } } ``` {% /tab %} {% /tabs %} {% /tab %} {% tab label="400" %} Bad Request {% tabs %} {% tab label="Model" %} API error response. | Parent field | Field | Type | Description | | ------------ | ------------------- | -------- | ------------------------------------------------------------------------------- | | | errors [*required*] | [object] | A list of errors. | | errors | detail | string | A human-readable explanation specific to this occurrence of the error. | | errors | meta | object | Non-standard meta-information about the error | | errors | source | object | References to the source of the error. | | source | header | string | A string indicating the name of a single request header which caused the error. | | source | parameter | string | A string indicating which URI query parameter caused the error. | | source | pointer | string | A JSON pointer to the value in the request document that caused the error. | | errors | status | string | Status code of the response. | | errors | title | string | Short human-readable summary of the error. | {% /tab %} {% tab label="Example" %} ```json { "errors": [ { "detail": "Missing required attribute in body", "meta": {}, "source": { "header": "Authorization", "parameter": "limit", "pointer": "/data/attributes/title" }, "status": "400", "title": "Bad Request" } ] } ``` {% /tab %} {% /tabs %} {% /tab %} {% tab label="401" %} Unauthorized {% tabs %} {% tab label="Model" %} API error response. | Parent field | Field | Type | Description | | ------------ | ------------------- | -------- | ------------------------------------------------------------------------------- | | | errors [*required*] | [object] | A list of errors. | | errors | detail | string | A human-readable explanation specific to this occurrence of the error. | | errors | meta | object | Non-standard meta-information about the error | | errors | source | object | References to the source of the error. | | source | header | string | A string indicating the name of a single request header which caused the error. | | source | parameter | string | A string indicating which URI query parameter caused the error. | | source | pointer | string | A JSON pointer to the value in the request document that caused the error. | | errors | status | string | Status code of the response. | | errors | title | string | Short human-readable summary of the error. | {% /tab %} {% tab label="Example" %} ```json { "errors": [ { "detail": "Missing required attribute in body", "meta": {}, "source": { "header": "Authorization", "parameter": "limit", "pointer": "/data/attributes/title" }, "status": "400", "title": "Bad Request" } ] } ``` {% /tab %} {% /tabs %} {% /tab %} {% tab label="403" %} Forbidden {% tabs %} {% tab label="Model" %} API error response. | Parent field | Field | Type | Description | | ------------ | ------------------- | -------- | ------------------------------------------------------------------------------- | | | errors [*required*] | [object] | A list of errors. | | errors | detail | string | A human-readable explanation specific to this occurrence of the error. | | errors | meta | object | Non-standard meta-information about the error | | errors | source | object | References to the source of the error. | | source | header | string | A string indicating the name of a single request header which caused the error. | | source | parameter | string | A string indicating which URI query parameter caused the error. | | source | pointer | string | A JSON pointer to the value in the request document that caused the error. | | errors | status | string | Status code of the response. | | errors | title | string | Short human-readable summary of the error. | {% /tab %} {% tab label="Example" %} ```json { "errors": [ { "detail": "Missing required attribute in body", "meta": {}, "source": { "header": "Authorization", "parameter": "limit", "pointer": "/data/attributes/title" }, "status": "400", "title": "Bad Request" } ] } ``` {% /tab %} {% /tabs %} {% /tab %} {% tab label="404" %} Not Found {% tabs %} {% tab label="Model" %} API error response. | Parent field | Field | Type | Description | | ------------ | ------------------- | -------- | ------------------------------------------------------------------------------- | | | errors [*required*] | [object] | A list of errors. | | errors | detail | string | A human-readable explanation specific to this occurrence of the error. | | errors | meta | object | Non-standard meta-information about the error | | errors | source | object | References to the source of the error. | | source | header | string | A string indicating the name of a single request header which caused the error. | | source | parameter | string | A string indicating which URI query parameter caused the error. | | source | pointer | string | A JSON pointer to the value in the request document that caused the error. | | errors | status | string | Status code of the response. | | errors | title | string | Short human-readable summary of the error. | {% /tab %} {% tab label="Example" %} ```json { "errors": [ { "detail": "Missing required attribute in body", "meta": {}, "source": { "header": "Authorization", "parameter": "limit", "pointer": "/data/attributes/title" }, "status": "400", "title": "Bad Request" } ] } ``` {% /tab %} {% /tabs %} {% /tab %} {% tab label="429" %} Too many requests {% tabs %} {% tab label="Model" %} API error response. | Field | Type | Description | | ------------------- | -------- | ----------------- | | errors [*required*] | [string] | A list of errors. | {% /tab %} {% tab label="Example" %} ```json { "errors": [ "Bad Request" ] } ``` {% /tab %} {% /tabs %} {% /tab %} {% tab label="500" %} Internal Server Error {% tabs %} {% tab label="Model" %} API error response. | Parent field | Field | Type | Description | | ------------ | ------------------- | -------- | ------------------------------------------------------------------------------- | | | errors [*required*] | [object] | A list of errors. | | errors | detail | string | A human-readable explanation specific to this occurrence of the error. | | errors | meta | object | Non-standard meta-information about the error | | errors | source | object | References to the source of the error. | | source | header | string | A string indicating the name of a single request header which caused the error. | | source | parameter | string | A string indicating which URI query parameter caused the error. | | source | pointer | string | A JSON pointer to the value in the request document that caused the error. | | errors | status | string | Status code of the response. | | errors | title | string | Short human-readable summary of the error. | {% /tab %} {% tab label="Example" %} ```json { "errors": [ { "detail": "Missing required attribute in body", "meta": {}, "source": { "header": "Authorization", "parameter": "limit", "pointer": "/data/attributes/title" }, "status": "400", "title": "Bad Request" } ] } ``` {% /tab %} {% /tabs %} {% /tab %} {% /tabs %} ### Code Example {% tabs %} {% tab label="Curl" %} ```bash # Path parameters export experiment_id="3fd6b5e0-8910-4b1c-a7d0-5b84de329012" # Set your Datadog site and credentials export DD_SITE="datadoghq.com" # Or us3.datadoghq.com, us5.datadoghq.com, datadoghq.eu, ap1.datadoghq.com, ap2.datadoghq.com, uk1.datadoghq.com, ddog-gov.com, us2.ddog-gov.com # Use a Personal Access Token or Service Access Token export DD_BEARER_TOKEN="" curl -X GET "https://api.${DD_SITE}/api/v2/llm-obs/v3/experiments/${experiment_id}/events" \ -H "Accept: application/json" \ -H "Authorization: Bearer ${DD_BEARER_TOKEN}" ``` {% /tab %} {% tab label="Go" %} ```go // List events for an Agent Observability experiment returns "OK" response package main import ( "context" "encoding/json" "fmt" "os" "github.com/DataDog/datadog-api-client-go/v2/api/datadog" "github.com/DataDog/datadog-api-client-go/v2/api/datadogV2" ) func main() { ctx := datadog.NewDefaultContext(context.Background()) ctx = context.WithValue(ctx, datadog.ContextAccessToken, os.Getenv("DD_BEARER_TOKEN")) configuration := datadog.NewConfiguration() configuration.SetUnstableOperationEnabled("v2.ListLLMObsExperimentEvents", true) apiClient := datadog.NewAPIClient(configuration) api := datadogV2.NewAgentObservabilityApi(apiClient) resp, r, err := api.ListLLMObsExperimentEvents(ctx, "experiment_id", *datadogV2.NewListLLMObsExperimentEventsOptionalParameters()) if err != nil { fmt.Fprintf(os.Stderr, "Error when calling `AgentObservabilityApi.ListLLMObsExperimentEvents`: %v\n", err) fmt.Fprintf(os.Stderr, "Full HTTP response: %v\n", r) } responseContent, _ := json.MarshalIndent(resp, "", " ") fmt.Fprintf(os.Stdout, "Response from `AgentObservabilityApi.ListLLMObsExperimentEvents`:\n%s\n", responseContent) } ``` #### Instructions First [install the library and its dependencies](https://dnd37xeb23o8n.cloudfront.net/api/latest.md?code-lang=go) and then save the example to `main.go` and run following commands: ```bash DD_SITE="datadoghq.com" DD_BEARER_TOKEN="" go run "main.go" ``` {% /tab %} {% tab label="Java" %} ```java // List events for an Agent Observability experiment returns "OK" response import com.datadog.api.client.ApiClient; import com.datadog.api.client.ApiException; import com.datadog.api.client.v2.api.AgentObservabilityApi; import com.datadog.api.client.v2.model.LLMObsExperimentEventsV2Response; public class Example { public static void main(String[] args) { ApiClient defaultClient = ApiClient.getDefaultApiClient(); defaultClient.setAccessToken(System.getenv("DD_BEARER_TOKEN")); defaultClient.setUnstableOperationEnabled("v2.listLLMObsExperimentEvents", true); AgentObservabilityApi apiInstance = new AgentObservabilityApi(defaultClient); try { LLMObsExperimentEventsV2Response result = apiInstance.listLLMObsExperimentEvents("3fd6b5e0-8910-4b1c-a7d0-5b84de329012"); System.out.println(result); } catch (ApiException e) { System.err.println("Exception when calling AgentObservabilityApi#listLLMObsExperimentEvents"); System.err.println("Status code: " + e.getCode()); System.err.println("Reason: " + e.getResponseBody()); System.err.println("Response headers: " + e.getResponseHeaders()); e.printStackTrace(); } } } ``` #### Instructions First [install the library and its dependencies](https://dnd37xeb23o8n.cloudfront.net/api/latest.md?code-lang=java) and then save the example to `Example.java` and run following commands: ```bash DD_SITE="datadoghq.com" DD_BEARER_TOKEN="" java "Example.java" ``` {% /tab %} {% tab label="Python" %} ```python """ List events for an Agent Observability experiment returns "OK" response """ from os import environ from datadog_api_client import ApiClient, Configuration from datadog_api_client.v2.api.agent_observability_api import AgentObservabilityApi configuration = Configuration() configuration.access_token = environ["DD_BEARER_TOKEN"] configuration.unstable_operations["list_llm_obs_experiment_events"] = True with ApiClient(configuration) as api_client: api_instance = AgentObservabilityApi(api_client) response = api_instance.list_llm_obs_experiment_events( experiment_id="experiment_id", ) print(response) ``` #### Instructions First [install the library and its dependencies](https://dnd37xeb23o8n.cloudfront.net/api/latest.md?code-lang=python) and then save the example to `example.py` and run following commands: ```bash DD_SITE="datadoghq.com" DD_BEARER_TOKEN="" python3 "example.py" ``` {% /tab %} {% tab label="Ruby" %} ```ruby # List events for an Agent Observability experiment returns "OK" response require "datadog_api_client" DatadogAPIClient.configure do |config| config.access_token = ENV["DD_BEARER_TOKEN"] config.unstable_operations["v2.list_llm_obs_experiment_events".to_sym] = true end api_instance = DatadogAPIClient::V2::AgentObservabilityAPI.new p api_instance.list_llm_obs_experiment_events("experiment_id") ``` #### Instructions First [install the library and its dependencies](https://dnd37xeb23o8n.cloudfront.net/api/latest.md?code-lang=ruby) and then save the example to `example.rb` and run following commands: ```bash DD_SITE="datadoghq.com" DD_BEARER_TOKEN="" rb "example.rb" ``` {% /tab %} {% tab label="Rust" %} ```rust // List events for an Agent Observability experiment returns "OK" response use datadog_api_client::datadog; use datadog_api_client::datadogV2::api_agent_observability::AgentObservabilityAPI; use datadog_api_client::datadogV2::api_agent_observability::ListLLMObsExperimentEventsOptionalParams; #[tokio::main] async fn main() { let mut configuration = datadog::Configuration::new(); configuration.set_unstable_operation_enabled("v2.list_llm_obs_experiment_events", true); let api = AgentObservabilityAPI::with_config(configuration); let resp = api .list_llm_obs_experiment_events( "experiment_id".to_string(), ListLLMObsExperimentEventsOptionalParams::default(), ) .await; if let Ok(value) = resp { println!("{:#?}", value); } else { println!("{:#?}", resp.unwrap_err()); } } ``` #### Instructions First [install the library and its dependencies](https://dnd37xeb23o8n.cloudfront.net/api/latest.md?code-lang=rust) and then save the example to `src/main.rs` and run following commands: ```bash DD_SITE="datadoghq.com" DD_API_KEY="" DD_APP_KEY="" cargo run ``` {% /tab %} {% tab label="TypeScript" %} ```typescript /** * List events for an Agent Observability experiment returns "OK" response */ import { client, v2 } from "@datadog/datadog-api-client"; const configuration = client.createConfiguration({ authMethods: { AuthZ: { accessToken: process.env.DD_BEARER_TOKEN as string, }, }, }); configuration.unstableOperations["v2.listLLMObsExperimentEvents"] = true; const apiInstance = new v2.AgentObservabilityApi(configuration); const params: v2.AgentObservabilityApiListLLMObsExperimentEventsRequest = { experimentId: "experiment_id", }; apiInstance .listLLMObsExperimentEvents(params) .then((data: v2.LLMObsExperimentEventsV2Response) => { console.log( "API called successfully. Returned data: " + JSON.stringify(data) ); }) .catch((error: any) => console.error(error)); ``` #### Instructions First [install the library and its dependencies](https://dnd37xeb23o8n.cloudfront.net/api/latest.md?code-lang=typescript) and then save the example to `example.ts` and run following commands: ```bash DD_SITE="datadoghq.com" DD_BEARER_TOKEN="" tsc "example.ts" ``` {% /tab %} {% /tabs %}