n8n/packages/@n8n/instance-ai/evaluations/computer-use/data/5.1-form-trigger-fill.json
Bernhard Wittmann b445221c6a
feat: Computer-use evaluation harness (no-changelog) (#29797)
Co-authored-by: Elias Meire <elias@meire.dev>
2026-05-12 08:36:12 +00:00

34 lines
1.1 KiB
JSON
Raw Permalink Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

{
"id": "5.1-form-trigger-fill",
"category": "browser",
"prompt": "I have an active workflow called 'CU Eval — Form Trigger Test' that has a Form trigger. Open its production form URL and fill it out with test data (name 'Test User', email 'test@example.com') so I can see if the workflow runs.",
"setup": {
"seedWorkflow": "form-trigger-workflow.json",
"activateSeededWorkflow": true
},
"budgets": { "maxToolCalls": 50, "maxDurationMs": 600000 },
"graders": [
{ "type": "trace.mustCallMcpServer", "server": "computer-use" },
{ "type": "trace.mustCallTool", "name": "browser_type" },
{ "type": "trace.mustNotLoop", "maxRepeatedCall": 4 },
{
"type": "trace.budget",
"maxToolCalls": 50,
"maxToolResultTokensEst": 200000,
"maxSingleToolResultTokensEst": 50000
},
{
"type": "trace.finalTextMatches",
"anyOf": ["submitted|filled|test user"],
"mustNotMatch": [
"taking a while",
"let me try (a )?different",
"couldn[']t (load|connect|reach)",
"timed? out",
"unable to (load|access|reach)"
]
}
],
"tags": ["browser", "form", "requires:browser-bootstrap"]
}