Test 173

Sep 14, 2026 / 1 fact

Request

Write slide 9 of Archived_2023_B_Round_Fundraising_v2 as an archive of the metrics-section structure I was considering on February 3, 2023. Record the intended metrics and customer-proof format without inventing historical numbers.

Required memory

Fact 8

Morgan Chen is leaning toward a tighter fundraising-deck metrics section covering net new ARR, retention shape, expansion by cohort, and one customer slide using actual customer names rather than anonymous logos.

Source evidence (1)

000096Feb 3, 2023 / 10:00 UTC-08:00

ok so B stuff is finally in a real state. deck v2 is decent now, not done-done but the story tracks and the ugly chart problems are mostly down to 2 slides instead of 8. Devon's notes helped. i've got ~8 intro calls booked for early Feb, mostly where i wanted them, and the lead-firm meeting is locked for week 2. still think partner-level chemistry matters way more than people admit at this stage, so i'm treating the first calls as filtering not just pitching. also trying not to let us wander into enterprise-AI wallpaper language in the deck -- if we say "developer workflow" one more vague time i'll delete the whole page lol. leaning keep the metrics section tighter: net new ARR, retention shape, expansion by cohort, then one customer slide with actual names and not anonymous logo soup. that's probably enough.

Message 000096 in history

Expected tool calls

update_deck_slide

Grading

1. field_equals / update_deck_slide

{
  "type": "field_equals",
  "tool": "update_deck_slide",
  "path": "result.ok",
  "value": true,
  "check_id": "morgan_173_00",
  "action_id": "morgan_173_update_deck_slide"
}

2. field_equals / update_deck_slide

{
  "type": "field_equals",
  "path": "result.deck",
  "value": "Archived_2023_B_Round_Fundraising_v2",
  "tool": "update_deck_slide",
  "check_id": "morgan_173_01",
  "action_id": "morgan_173_update_deck_slide"
}

3. field_eq_number / update_deck_slide

{
  "type": "field_eq_number",
  "path": "result.slide.number",
  "value": 9,
  "tool": "update_deck_slide",
  "check_id": "morgan_173_02",
  "action_id": "morgan_173_update_deck_slide"
}

4. field_llm_judge / update_deck_slide

{
  "type": "field_llm_judge",
  "path": "result.slide.content",
  "criterion": "Records the contemplated structure as net new ARR, retention shape, expansion by cohort, then one customer slide with actual names rather than anonymous logos. Does not add unrelated core metric categories, invent numbers, or claim the tentative structure was definitively approved.",
  "tool": "update_deck_slide",
  "check_id": "morgan_173_03",
  "action_id": "morgan_173_update_deck_slide"
}

Complete grading specification

{
  "type": "tool_trace",
  "config": {
    "check_version": 2,
    "today": "2026-09-14",
    "assertions": [
      {
        "type": "field_equals",
        "tool": "update_deck_slide",
        "path": "result.ok",
        "value": true,
        "check_id": "morgan_173_00",
        "action_id": "morgan_173_update_deck_slide"
      },
      {
        "type": "field_equals",
        "path": "result.deck",
        "value": "Archived_2023_B_Round_Fundraising_v2",
        "tool": "update_deck_slide",
        "check_id": "morgan_173_01",
        "action_id": "morgan_173_update_deck_slide"
      },
      {
        "type": "field_eq_number",
        "path": "result.slide.number",
        "value": 9,
        "tool": "update_deck_slide",
        "check_id": "morgan_173_02",
        "action_id": "morgan_173_update_deck_slide"
      },
      {
        "type": "field_llm_judge",
        "path": "result.slide.content",
        "criterion": "Records the contemplated structure as net new ARR, retention shape, expansion by cohort, then one customer slide with actual names rather than anonymous logos. Does not add unrelated core metric categories, invent numbers, or claim the tentative structure was definitively approved.",
        "tool": "update_deck_slide",
        "check_id": "morgan_173_03",
        "action_id": "morgan_173_update_deck_slide"
      }
    ]
  }
}

App state

Download JSON

Source file

tests/morgan/173.yaml

SHA-256: 6f08e2fdbde71c40529326dab34990f6b69af360f08824307133c9802e187e47