Evaluate Run

Complete a run and evaluate against test assertions

POST /api/platform/evaluateRun

Evaluate Run

cURL

curl --request POST \
  --url https://api.example.com/api/platform/evaluateRun \
  --header 'Content-Type: application/json' \
  --data '
{
  "runId": "<string>"
}
'

import requests

url = "https://api.example.com/api/platform/evaluateRun"

payload = { "runId": "<string>" }
headers = {"Content-Type": "application/json"}

response = requests.post(url, json=payload, headers=headers)

print(response.text)

const options = {
  method: 'POST',
  headers: {'Content-Type': 'application/json'},
  body: JSON.stringify({runId: '<string>'})
};

fetch('https://api.example.com/api/platform/evaluateRun', options)
  .then(res => res.json())
  .then(res => console.log(res))
  .catch(err => console.error(err));

<?php

$curl = curl_init();

curl_setopt_array($curl, [
  CURLOPT_URL => "https://api.example.com/api/platform/evaluateRun",
  CURLOPT_RETURNTRANSFER => true,
  CURLOPT_ENCODING => "",
  CURLOPT_MAXREDIRS => 10,
  CURLOPT_TIMEOUT => 30,
  CURLOPT_HTTP_VERSION => CURL_HTTP_VERSION_1_1,
  CURLOPT_CUSTOMREQUEST => "POST",
  CURLOPT_POSTFIELDS => json_encode([
    'runId' => '<string>'
  ]),
  CURLOPT_HTTPHEADER => [
    "Content-Type: application/json"
  ],
]);

$response = curl_exec($curl);
$err = curl_error($curl);

curl_close($curl);

if ($err) {
  echo "cURL Error #:" . $err;
} else {
  echo $response;
}

package main

import (
	"fmt"
	"strings"
	"net/http"
	"io"
)

func main() {

	url := "https://api.example.com/api/platform/evaluateRun"

	payload := strings.NewReader("{\n  \"runId\": \"<string>\"\n}")

	req, _ := http.NewRequest("POST", url, payload)

	req.Header.Add("Content-Type", "application/json")

	res, _ := http.DefaultClient.Do(req)

	defer res.Body.Close()
	body, _ := io.ReadAll(res.Body)

	fmt.Println(string(body))

}

HttpResponse<String> response = Unirest.post("https://api.example.com/api/platform/evaluateRun")
  .header("Content-Type", "application/json")
  .body("{\n  \"runId\": \"<string>\"\n}")
  .asString();

require 'uri'
require 'net/http'

url = URI("https://api.example.com/api/platform/evaluateRun")

http = Net::HTTP.new(url.host, url.port)
http.use_ssl = true

request = Net::HTTP::Post.new(url)
request["Content-Type"] = 'application/json'
request.body = "{\n  \"runId\": \"<string>\"\n}"

response = http.request(request)
puts response.read_body

200

{
  "runId": "<string>",
  "status": "<string>",
  "passed": true,
  "score": {},
  "failures": [
    {}
  ]
}

Request

POST /api/platform/evaluateRun

Body Parameters

runId · string · required

Run ID from startRun

Example Request

curl -X POST https://api.agentdiff.dev/api/platform/evaluateRun \
  -H "X-API-Key: ad_live_sk_..." \
  -H "Content-Type: application/json" \
  -d '{
    "runId": "run-xyz789"
  }'

Response

runId · string

Run identifier

status · string

Final status: "passed" or "failed"

passed · boolean

Whether all assertions passed

score · object

Score details

failures · array

List of failed assertions with details

Example Response (Passed)

{
  "runId": "run-xyz789",
  "status": "passed",
  "passed": true,
  "score": {
    "passed": 2,
    "total": 2,
    "percent": 100.0
  },
  "failures": []
}

Example Response (Failed)

{
  "runId": "run-xyz789",
  "status": "failed",
  "passed": false,
  "score": {
    "passed": 1,
    "total": 2,
    "percent": 50.0
  },
  "failures": [
    "assertion#2 messages expected count 1 but got 0"
  ]
}

What Happens

1. After snapshot: Current database state is captured
2. Diff computed: Before/after states compared
3. Assertions evaluated: Test assertions checked against diff
4. Results stored: Results saved for later retrieval
5. Replication stopped: WAL capture ends

Use this when you have a test with assertions. For just getting the diff without assertions, use diffRun instead.

Errors

Error	Status	Description
run_not_found	404	Run doesn’t exist
run_already_completed	400	Run already evaluated
no_test_defined	400	No test associated with this run

SDK Usage

Python

result = client.evaluate_run(runId=run.runId)

print(f"Passed: {result.passed}")
print(f"Score: {result.score['percent']}%")
if result.failures:
    for f in result.failures:
        print(f"  Failed: {f}")

const result = await client.evaluateRun({ runId: run.runId });

console.log(`Passed: ${result.passed}`);
console.log(`Score: ${result.score.percent}%`);
if (result.failures) {
  result.failures.forEach(f => console.log(`  Failed: ${f}`));
}