Voxli Voxli

Test Results

Update test result

Update an existing test result's properties such as external links, metadata, or status. A failed or canceled result is final: changing its status is rejected with 409.

PATCH /test-results/{id}

Authorization

HeaderTypeDescription
Authorizationrequired
string
Header with your API key as a Bearer token: Bearer <api-key>. See Authentication.

Path parameters

NameTypeDescription
idrequired
string
The id of the item to update

Body

application/json
PropertyTypeDescription
externalLinks
object[]
Optional external links (max 10)
namerequired
string
Display name for the link
urlrequired
string (uri)
URL to external resource
metadata
object
Optional metadata
status
enum<string>
Test result status
Allowed values: pending, running, completed, failed, canceled
completionMode
enum<string>
Change how the result completes. Must be set to 'manual' before the conversation ends to hold completion for POST /test-results/{id}/finish.
Allowed values: auto, manual
externalId
string
The platform's own id for this conversation, e.g. its chat id. A hosted tool endpoint links the session with this key, and its tool calls, to this test result.

Response

  • 200Successful Response
PropertyTypeDescription
idrequired
string
createdAt
string (date-time)
updatedAt
string (date-time)
test
object
id
string
name
string
run
object
id
string
agent
object
id
string
name
string
personality
object
The personality of the simulated user
id
string
name
string
statusrequired
enum<string>
Allowed values: pending, running, completed, failed, canceled
isEndedrequired
boolean
Whether the conversation has ended
endReason
enum<string>
Why the conversation ended, once it has
Allowed values: AGENT, MAX_TURNS_REACHED
completionModerequired
enum<string>
'auto' completes the result (processed and scored) as soon as the conversation ends. 'manual' waits for POST /test-results/{id}/finish so the client can attach data after the chat. Server-driven agents always complete automatically.
Allowed values: auto, manual
error
string
Why the conversation could not run, when status is failed
conversation
object[]
Default: []
id
string
Server-generated identifier used to address this message, e.g. when attaching metadata after the fact.
role
enum<string>
user is the simulated user (Voxli's tester), assistant the agent under test.
Allowed values: user, assistantDefault: "assistant"
type
enum<string>
Allowed values: message, tool, internal-event, public-event, actionDefault: "message"
content
string
name
string
metadata
object
Optional user-defined metadata for tool calls. Can contain any JSON-serializable data (max 100KB).
createdAt
string (date-time)
UTC time the message was added to the conversation.
assertionResults
object[]
Default: []
criteriarequired
string
The assertion criteria that was evaluated
severityrequired
enum<string>
The severity level of this assertion
Allowed values: blocker, medium, low
passed
boolean
Whether the assertion passed or failed
explanation
string
Explanation of why the assertion passed or failed
relatedMessageIndices
integer[]
0-indexed positions of messages relevant to this assertion
Default: []
scorerequired
integer
Weighted score based on assertion results
metadata
object
Metadata attached by your integration, e.g. the platform's chatId
Default: {}
externalLinks
object[]
Default: []
namerequired
string
Display name for the link
urlrequired
string (uri)
URL to external resource
externalId
string
The platform's own id for this conversation, e.g. its chat id

Test Results

Retry test result