Skip to main content
PUT
Update an evaluation case
This operation replaces an existing evaluation case; it is not a partial update. Retrieve and preserve any content you need before submitting the complete replacement
A nonempty request evalId must match the path eval_case_id. Conflicting IDs return 400, a missing target returns 404, and success returns JSON null

Authorizations

Authorization
string
header
required

Optional locally without a gateway; cloud deployments use the Runtime API key or user-pool JWT required by that deployment, never the model API key

Path Parameters

app_name
string
required

Application name; harness_agent for the CLI deployment

eval_set_id
string
required

Evaluation set ID

eval_case_id
string
required

Evaluation case ID

Body

application/json
evalId
string
required

Evaluation case ID

conversation
Invocation · object[] | null

Conversation invocations used in the evaluation

conversationScenario
ConversationScenario · object | null
sessionInput
SessionInput · object | null

Application, user, and initial state for the evaluation session

creationTimestamp
number
default:0

Creation time as a Unix timestamp in seconds

rubrics
Rubric · object[] | null
finalSessionState
Finalsessionstate · object | null

Expected final session state

Response

Successful response

Last modified on September 19, 2026