app_verdict
This tool on every surface: CLI · MCP · Python · JavaScript · Go · Java · .NET
Give a verdict on a claim about a recording, and have it held to the rules app_check holds a model to: status passed, failed or inconclusive; a summary; at most 12 observations; and evidence for passed or failed that cites at least one call the recording holds, by id (call@3). Inconclusive when the recording cannot settle the claim — missing evidence is never success. Read the recording with app_recording first. Answers the verdict as accepted, or why it was not. Needs no device and no credential.
Read-only: readOnlyHint is true.
Arguments
| Name | Type | Required | What it is |
|---|---|---|---|
claim | string | yes | The claim judged. |
evidence | array of object | yes | |
record | string | yes | The recording it was judged on. |
status | string: passed, failed, inconclusive | yes | |
summary | string | yes | The verdict in a sentence or two. |
Examples
Each is the arguments of a tools/call request; the first is shown whole.
Judge a claim yourself, citing the call it rests on
{
"jsonrpc": "2.0",
"id": 1,
"method": "tools/call",
"params": {
"name": "app_verdict",
"arguments": {
"claim": "After card 3 was tapped, it was shown face up as the Blue Morpho.",
"record": "card.zip",
"status": "passed",
"summary": "Card 3 reads Blue Morpho after the tap",
"evidence": [
{
"type": "observation",
"summary": "call@7: the screen after the tap shows Card 3, Blue Morpho"
}
]
}
}
}