Read a conversation test run

GET /v1/agents/{agent_id}/versions/{version}/tests/{run_id}

Answers the report of a conversation test run (1.55.0): per test and repetition, each expectation met or not with the count that decided it, the Call by id, its ended reason and Turn count. Ids, codes and counts only, so the read is not audited. From 1.64.0, stated: each Call the report names, of inline tests or of a suite, is a test Call read through get_call and its Turns and text through get_call_transcript, an audited content read, as for any test Call; get_prompt_eval_call reads the Calls of the runs of a suite, and an inline test, having no suite, has no other route.

Request

Set VOCAL_API to the base URL of your API and VOCAL_API_KEY to an API key with the prompts:read scope (Authentication and keys). Replace each placeholder, such as agt_... or <name>, with your own value.

curl "$VOCAL_API/v1/agents/agt_.../versions/1/tests/per_..." \
  -H "Authorization: Bearer $VOCAL_API_KEY"

const response = await fetch(`${process.env.VOCAL_API}/v1/agents/agt_.../versions/1/tests/per_...`, {
  headers: {
    Authorization: `Bearer ${process.env.VOCAL_API_KEY}`,
  },
});
console.log(response.status, await response.text());

import os

import requests

response = requests.get(
    f"{os.environ['VOCAL_API']}/v1/agents/agt_.../versions/1/tests/per_...",
    headers={
        "Authorization": f"Bearer {os.environ['VOCAL_API_KEY']}",
    },
)
print(response.status_code, response.text)

Parameters

Responses

Fields of the 200 answer

The shape is ConversationTestRun.