Test Runs

Each execution of a test plan creates a test run. Runs record individual test results, agent outputs, screenshots, and logs.

Run lifecycle

Status	Description
Queued	Run accepted and waiting for execution infrastructure
Running	Tests are actively executing; results stream in real time
Analysing	Tests complete; post-run agents are processing results
Passed	All tests passed; no regressions detected
Failed	One or more tests failed; see results for details
Error	Infrastructure error prevented the run from completing; all results may not be available

Viewing run results

Navigate to Test Runs → [run name]. The results view has tabs:

Summary — pass count, fail count, duration, environment, triggered by
Tests — per-test results with status, duration, assertion details
Agent Activity — outputs from all post-run agents (ResultAnalyzer, RegressionGuard, etc.)
Coverage — CoverageGap report for this run
Logs — raw test execution logs per test
Screenshots — failure screenshots for Web UI tests

Per-test detail panel

Click the Tests (N) expander on any run card to open the per-test detail panel without leaving the runs list. Each test row shows:

Outcome dot (green = pass, red = fail, amber = flaky, grey = skipped)
Artifact badges indicating which artifacts are available: screenshot, browser console log, API request/response log, Playwright trace

Clicking a test row expands the full detail view, which loads on demand:

Traceback — full Python exception and assertion error text
Stdout — captured print statements and test step logs
Screenshot thumbnail — inline failure screenshot for Web UI tests; click to open full size
Browser console log — JavaScript console output captured during the test session
API request/response log — full curl-equivalent request plus response body for API tests; includes headers, payload, status code, and elapsed milliseconds
Playwright trace download — download the .zip trace file for replay in the Playwright Trace Viewer

The API log captures all HTTP calls made during the test, including auth requests, so you can reproduce failures with a single curl command without re-running the suite.

Retrying failed tests

From the Tests tab, select one or more failed tests and click Re-run Selected to create a new run containing only those tests. This is useful for verifying a fix without running the full suite.

Run retention

Run results are retained for:

Pro plan — 90 days
Enterprise plan — configurable, up to unlimited

After the retention period, detailed logs and screenshots are deleted. Summary-level data (pass/fail counts, run metadata) is retained indefinitely for trend reporting.

Run API

Check run status programmatically to integrate with CI/CD gates:

GET https://api.sentinelflux.in/v1/runs/{run_id}
Authorization: Bearer {API_TOKEN}

# Response
{
  "run_id": "run_abc123",
  "status": "failed",
  "passed": 34,
  "failed": 2,
  "regression_guard_status": "regressions_detected",
  "duration_seconds": 187,
  "completed_at": "2025-11-15T14:23:01Z"
}