Complete grading specification
{
"type": "tool_trace",
"config": {
"check_version": 2,
"today": "2028-01-01",
"semantic_judge_version": 2,
"assertions": [
{
"type": "field_equals",
"tool": "send_slack_dm",
"action_id": "dm_owen",
"path": "result.ok",
"value": true,
"check_id": "riley_084_00"
},
{
"type": "field_equals",
"tool": "send_slack_dm",
"action_id": "dm_ines",
"path": "result.ok",
"value": true,
"check_id": "riley_084_01"
},
{
"type": "field_equals",
"tool": "send_slack_dm",
"action_id": "dm_daniela",
"path": "result.ok",
"value": true,
"check_id": "riley_084_02"
},
{
"type": "field_equals",
"tool": "send_slack_dm",
"action_id": "dm_owen",
"path": "args.user",
"value": "owen",
"check_id": "riley_084_03"
},
{
"type": "field_equals",
"tool": "send_slack_dm",
"action_id": "dm_ines",
"path": "args.user",
"value": "ines",
"check_id": "riley_084_04"
},
{
"type": "field_equals",
"tool": "send_slack_dm",
"action_id": "dm_daniela",
"path": "args.user",
"value": "daniela",
"check_id": "riley_084_05"
},
{
"type": "field_llm_judge",
"tool": "send_slack_dm",
"action_id": "dm_owen",
"path": "args",
"criterion": "The message is a concise request for Owen to confirm by January 5, 2028 that his established shared-state lane is reporting semantics. It must remain limited to Owen's lane and must not assign or imply that he has authority over messaging consumption or transport/source reliability.",
"check_id": "riley_084_06"
},
{
"type": "field_llm_judge",
"tool": "send_slack_dm",
"action_id": "dm_ines",
"path": "args",
"criterion": "The message is a concise request for Ines to confirm by January 5, 2028 that her established shared-state lane is how messaging consumes the shared states. It must remain limited to Ines's lane and must not assign or imply that she has authority over reporting semantics or transport/source reliability.",
"check_id": "riley_084_07"
},
{
"type": "field_llm_judge",
"tool": "send_slack_dm",
"action_id": "dm_daniela",
"path": "args",
"criterion": "The message is a concise request for Daniela to confirm by January 5, 2028 that her established shared-state lane is transport and source reliability. It must remain limited to Daniela's lane and must not assign or imply that she has authority over reporting semantics or messaging consumption.",
"check_id": "riley_084_08"
}
]
}
}