Test 120

Jan 1, 2028 / 1 fact

Request

Create a compact archival ownership card titled “July 2023 analysis ownership card” for the Operating Archive. Cover the Helio Start baseline analysis and retained-mid expansion eligibility analysis as two distinct lanes, identify the default owner for each, and make clear that these assignments were limited to July 2023 rather than a current standing org rule.

Required memory

Fact 408

For July 2023, Owen is the default owner of the Helio Start baseline readout shape and the retained-mid expansion eligibility cuts.

Source evidence (1)

000377Jul 4, 2023 / 11:38 UTC-05:00

Quick note to remember for later: the 10:30 owner-lanes block with Priya finished, and we did make a concrete July split, but not as a permanent org chart. Owen is the default owner of the Helio Start baseline readout shape and the retained-mid expansion eligibility cuts. Ines owns the lifecycle fork risk map and the copy boundaries that flow from both Helio Start and the expansion-nudge learning. I keep the decision rules, executive tradeoffs, and cross-functional guardrails. Priya was explicit that we are reusing the investigation framework informally for July work, not turning it into a formal company process. No deliverable right now; I just want that captured.

Message 000377 in history

Expected tool calls

create_doc

Grading

1. field_equals / create_doc

{
  "type": "field_equals",
  "tool": "create_doc",
  "action_id": "create_july_ownership_card",
  "path": "result.ok",
  "value": true,
  "check_id": "riley_120_00"
}

2. field_equals / create_doc

{
  "check_id": "riley_120_01",
  "type": "field_equals",
  "tool": "create_doc",
  "action_id": "create_july_ownership_card",
  "path": "args.title",
  "value": "July 2023 analysis ownership card"
}

3. field_equals / create_doc

{
  "check_id": "riley_120_02",
  "type": "field_equals",
  "tool": "create_doc",
  "action_id": "create_july_ownership_card",
  "path": "args.folder",
  "value": "Operating Archive"
}

4. field_llm_judge / create_doc

{
  "type": "field_llm_judge",
  "tool": "create_doc",
  "action_id": "create_july_ownership_card",
  "path": "args",
  "criterion": "The created document must clearly function as an archival July 2023 ownership card and state that Owen was the default owner for each of two separate analysis lanes: (1) the Helio Start baseline readout shape and (2) the retained-mid expansion eligibility cuts. It must also clearly limit this assignment to July 2023 and avoid presenting it as a current or permanent org rule. Missing, merging, or contradicting either lane, owner, or time limitation must fail.",
  "check_id": "riley_120_03"
}

Complete grading specification

{
  "type": "tool_trace",
  "config": {
    "check_version": 2,
    "today": "2028-01-01",
    "semantic_judge_version": 2,
    "assertions": [
      {
        "type": "field_equals",
        "tool": "create_doc",
        "action_id": "create_july_ownership_card",
        "path": "result.ok",
        "value": true,
        "check_id": "riley_120_00"
      },
      {
        "check_id": "riley_120_01",
        "type": "field_equals",
        "tool": "create_doc",
        "action_id": "create_july_ownership_card",
        "path": "args.title",
        "value": "July 2023 analysis ownership card"
      },
      {
        "check_id": "riley_120_02",
        "type": "field_equals",
        "tool": "create_doc",
        "action_id": "create_july_ownership_card",
        "path": "args.folder",
        "value": "Operating Archive"
      },
      {
        "type": "field_llm_judge",
        "tool": "create_doc",
        "action_id": "create_july_ownership_card",
        "path": "args",
        "criterion": "The created document must clearly function as an archival July 2023 ownership card and state that Owen was the default owner for each of two separate analysis lanes: (1) the Helio Start baseline readout shape and (2) the retained-mid expansion eligibility cuts. It must also clearly limit this assignment to July 2023 and avoid presenting it as a current or permanent org rule. Missing, merging, or contradicting either lane, owner, or time limitation must fail.",
        "check_id": "riley_120_03"
      }
    ]
  }
}

App state

Download JSON

Source file

tests/riley/120.yaml

SHA-256: 57f311e6b7fc48ea14f242a14754a0dad733f13a279fad7e59b1ae7d5fee38a1