Test 164

Sep 14, 2026 / 2 facts

Request

Create a May 2023 Mercury planning archive note explaining where operating work was coordinated and how investor-deck work was prioritized.

Required memory

Fact 102

Morgan Chen’s May 4 Friday-wrap notes report that launch work is being tightened before the broader investor story is reopened.

Source evidence (1)

000476May 5, 2023 / 08:22 UTC-07:00

Here's the rough Friday wrap. Notes — Friday wrap rough May 4 team - shipped the standing-rules pass. - agent workflow is now actually in use, not just a draft / runbook thing. - keep this as movement, not a big operating manifesto. customers - product moved, but the blockers are still showing up in the same places. - activation still has a dead-end around invite/setup for some new accounts. - admin invite failures still unresolved in a couple of account shapes. - account-level permissions are still muddy enough to block rollout / confuse admins. - keep this plain. no spin. investors - deck is still deferred. - tightening launch work first instead of reopening the whole story. - one line only. legal/ops - no decision this week, so leave it out. overall - should read like a short internal update, not a board note. - compact. Turn it into a compact internal draft I can review. Keep it as an update, not a planning artifact.

Message 000476 in history

Fact 103

On May 5, 2023, Morgan Chen requested a short handoff to Devon Hayes and Jake stating that Jake’s weekly is where Anna’s retention work, activation blockers, and admin blockers are handled, and that no investor material should be routed to Jake because the weekly is for product and operating work only.

Source evidence (1)

000478May 5, 2023 / 10:18 UTC-07:00

ok wrap is posted. Devon and I need to stop trying to solve this in the Friday wrap. Create/save a Mercury weekly operating note and send Devon + Jake a short handoff: Jake's weekly is where Anna's retention work, activation blockers, and admin blockers get handled. No investor material routed to Jake -- product/operating only.

Message 000478 in history

Expected tool calls

create_doc

Grading

1. field_equals / create_doc

{
  "type": "field_equals",
  "tool": "create_doc",
  "path": "result.ok",
  "value": true,
  "check_id": "morgan_164_00",
  "action_id": "morgan_164_create_doc"
}

2. field_llm_judge / create_doc

{
  "type": "field_llm_judge",
  "path": "result.document.body",
  "criterion": "Places retention, activation, and admin blockers in Jake's weekly, with investor material kept out of Jake's operating work. Keeps deck work deferred behind tightening launch work rather than reopening the investor story. No separate retention-evidence prerequisite for the deck is required.",
  "tool": "create_doc",
  "check_id": "morgan_164_01",
  "action_id": "morgan_164_create_doc"
}

Complete grading specification

{
  "type": "tool_trace",
  "config": {
    "check_version": 2,
    "today": "2026-09-14",
    "assertions": [
      {
        "type": "field_equals",
        "tool": "create_doc",
        "path": "result.ok",
        "value": true,
        "check_id": "morgan_164_00",
        "action_id": "morgan_164_create_doc"
      },
      {
        "type": "field_llm_judge",
        "path": "result.document.body",
        "criterion": "Places retention, activation, and admin blockers in Jake's weekly, with investor material kept out of Jake's operating work. Keeps deck work deferred behind tightening launch work rather than reopening the investor story. No separate retention-evidence prerequisite for the deck is required.",
        "tool": "create_doc",
        "check_id": "morgan_164_01",
        "action_id": "morgan_164_create_doc"
      }
    ]
  }
}

App state

Download JSON

Source file

tests/morgan/164.yaml

SHA-256: 0906253600446e06eeb63989db64d6a71f1347758e061402588e42a759460944