/api/v1/executions

GET

/api/v1/executions

const url = 'https://api.endless.zaia.app/api/v1/executions?pageNumber=1&pageSize=10';
const options = {method: 'GET'};

try {
  const response = await fetch(url, options);
  const data = await response.json();
  console.log(data);
} catch (error) {
  console.error(error);
}

curl --request GET \
  --url 'https://api.endless.zaia.app/api/v1/executions?pageNumber=1&pageSize=10'

Retrieves multiple executions.

Parameters

Query Parameters

pageNumber

The page number.

number

default: 1 >= 1

The page number.

pageSize

The page size.

number

default: 10 >= 1 <= 100

The page size.

responderIds

Any of:

Array<string>
string

Array<string>

Responses

200

Retrieves all executions for the workspace.

application/json

Array<object>

object

createdAt

required

string format: date-time

required

A UUID v4 string that identifies an entity.

string format: uuid

updatedAt

required

string format: date-time

creditsUsed

required

The total number of credits used by this execution.

number

detail

required

Any of:

object

input

required

The input data for this execution.

object

key

additional properties

nullable

output

required

The output data for this execution.

object

key

additional properties

nullable

agentId

required

The agent identifier.

string

agentVersionId

The exact agent version identifier used by this execution.

string

events

required

The events that occurred during this agent execution.

Array

Any of:

object
object

object

contextUsage

The context usage snapshot recorded for a response breakpoint.

object

categories

required

The structural character breakdown of the context.

object

conversation

required

The estimated conversation character count.

integer

mcpDefinitions

required

The estimated MCP definition character count.

integer

summarizedConversation

required

The summarized conversation character count.

integer

systemPrompt

required

The system prompt character count.

integer

toolCalls

required

The tool call character count.

integer

toolDefinitions

required

The tool definition character count.

integer

toolResults

required

The tool result character count.

integer

compaction

required

The tool result compaction summary.

object

afterCharacters

required

The tool result character count after compaction.

integer

beforeCharacters

required

The tool result character count before compaction.

integer

compactedResults

required

The number of tool results compacted.

integer

slicedResults

required

The number of tool results sliced as an overflow fallback.

integer

contextLimitTokens

required

The model context window limit in tokens.

integer

>= 1

fallbackReason

Why the prepared runtime fell back to legacy behavior.

string

Allowed values: low-confidence preparation-error provider-incompatible schema-invalid timeout

inputTokens

required

The input token count reported by the provider.

integer

nullable

model

required

The model used for this response.

string

outputReserveTokens

required

The output token reserve kept outside the input budget.

integer

recordedAt

required

When the context usage snapshot was recorded.

string format: date-time

runtime

required

The runtime that generated the response.

string

Allowed values: legacy prepared

toolSearchMode

The tool search mode used by the prepared runtime.

string

Allowed values: native-anthropic native-openai platform-search

creditsUsed

required

The number of credits used by this event.

number

endedAt

required

When this event ended.

string format: date-time

nullable

required

The event identifier.

string

input

Any of:

unknown
object

The input data for this event.

nullable

label

required

The label for this event.

string

>= 3 characters <= 50 characters

llmUsages

The custom-provider LLM usages recorded for this event.

Array<object>

object

callKind

The logical LLM call category.

string

Allowed values: agent compact load preparation repair replan search verify json-output

inputTokens

required

The input token count reported by the LLM provider.

integer

nullable

model

required

The model key used for this LLM usage.

string

outputTokens

required

The output token count reported by the LLM provider.

integer

nullable

provider

required

The provider metadata associated with this LLM usage.

object

required

The provider ID associated with the LLM usage, when one exists.

string

nullable

name

required

The display name of the provider associated with the LLM usage.

string

>= 3 characters <= 50 characters

runtime

The runtime that generated the LLM usage.

string

Allowed values: legacy prepared

output

required

The output data for this event.

object

key

additional properties

nullable

startedAt

required

When this event started.

string format: date-time

status

required

The status of the event.

string

Allowed values: running completed failed skipped

llm

required

The model key used during the iteration event.

string

llmProvider

required

The provider metadata associated with the iteration event.

object

required

The provider ID associated with the LLM usage, when one exists.

string

nullable

name

required

The display name of the provider associated with the LLM usage.

string

>= 3 characters <= 50 characters

type

required

The event subtype represented by this payload.

string

Allowed values: iteration

origin

required

Any of:

object

apiKeyId

required

The API key identifier.

string

type

required

The execution origin subtype represented by this payload.

string

Allowed values: api

responseBreakpoints

The context usage snapshots recorded for generated responses.

Array<object>

object

categories

required

The structural character breakdown of the context.

object

conversation

required

The estimated conversation character count.

integer

mcpDefinitions

required

The estimated MCP definition character count.

integer

summarizedConversation

required

The summarized conversation character count.

integer

systemPrompt

required

The system prompt character count.

integer

toolCalls

required

The tool call character count.

integer

toolDefinitions

required

The tool definition character count.

integer

toolResults

required

The tool result character count.

integer

compaction

required

The tool result compaction summary.

object

afterCharacters

required

The tool result character count after compaction.

integer

beforeCharacters

required

The tool result character count before compaction.

integer

compactedResults

required

The number of tool results compacted.

integer

slicedResults

required

The number of tool results sliced as an overflow fallback.

integer

contextLimitTokens

required

The model context window limit in tokens.

integer

>= 1

fallbackReason

Why the prepared runtime fell back to legacy behavior.

string

Allowed values: low-confidence preparation-error provider-incompatible schema-invalid timeout

inputTokens

required

The input token count reported by the provider.

integer

nullable

model

required

The model used for this response.

string

outputReserveTokens

required

The output token reserve kept outside the input budget.

integer

recordedAt

required

When the context usage snapshot was recorded.

string format: date-time

runtime

required

The runtime that generated the response.

string

Allowed values: legacy prepared

toolSearchMode

The tool search mode used by the prepared runtime.

string

Allowed values: native-anthropic native-openai platform-search

type

required

The execution detail subtype represented by this payload.

string

Allowed values: agent-execution

endedAt

required

When this execution ended.

string format: date-time

nullable

isAgentic

required

Whether this execution is agentic.

boolean

isFailedAttempt

required

Whether this execution belongs to a failed retry attempt.

boolean

isPurged

required

Whether this execution has been purged.

boolean

parentId

required

The parent execution ID of this execution.

string format: uuid

nullable

responderId

required

The responder ID of this execution.

string

startedAt

required

When this execution started.

string format: date-time

status

required

The status of the execution.

string

Allowed values: running completed failed cancelled

workspaceId

required

The workspace ID of this execution.

string

Example

[
  {
    "detail": {
      "events": [
        {
          "contextUsage": {
            "fallbackReason": "low-confidence",
            "runtime": "legacy",
            "toolSearchMode": "native-anthropic"
          },
          "llmUsages": [
            {
              "callKind": "agent",
              "runtime": "legacy"
            }
          ],
          "status": "running",
          "type": "iteration"
        }
      ],
      "origin": {
        "type": "api"
      },
      "responseBreakpoints": [
        {
          "fallbackReason": "low-confidence",
          "runtime": "legacy",
          "toolSearchMode": "native-anthropic"
        }
      ],
      "type": "agent-execution"
    },
    "status": "running"
  }
]

401

Unauthorized.

application/json

The description of a known exception.

object

code

The error code.

string

cta

The error call-to-action.

object

type

required

string

Allowed values: support-contact

url

required

string format: uri

feedback

required

The error feedback.

object

enUs

string

esEs

string

ptBr

string

message

required

The error message.

string

name

required

The error name.

string

Example

{
  "cta": {
    "type": "support-contact"
  }
}