CHECKER RUN 2026-09-07 09:19:11 | project NDIS and Aged Care Business Enabling Study | baseline v2.0.0 | AI Project Governance Standard v3.9 with the v4.0 Business Layer (7 September 2026)
===================================================================================================================================================================================================
C01 PASS GOV-B3.1, B3.6, B3.9 Every requirement has 11 non-empty fields, exactly one 'shall', no banned qualifier, and one of Inspection/Analysis/Demonstration/Test
32 requirements clean
C02 PASS GOV-F1.9, REQ-CON-05 Every assumption records a confidence level and what breaks if it is wrong
31 assumptions, all with confidence and a stated failure consequence
C03 PASS GOV-F1.16, REQ-CON-03 Every source has a resolvable URL and an accessed date inside the staleness window: not after the currency date, and not more than 90 days before it
74 sources, all with a URL and all inside the 90-day window against a currency date of 2026-09-07; the oldest was accessed 18 days before it
C04 PASS GOV-F8.2, REQ-SYS-18 Every source below High confidence states why it is below High
51 of 74 sources are below High; every one states why
C05 PASS GOV-F8.8, REQ-CON-01, REQ-MOP-01 Every material figure in the delivered study exists in the source register, the assumptions register, or the computed model (ratio >= 0.90)
sourced-claim ratio 0.968 over 309 material figures; 10 unmatched -> 123.0, 403.0, 406.39, 1849.0, 5325.0, 6797.0, 9997.0, 16200.0, 16900.0, 20959.0
C06 PASS GOV-F2.2 The dashboard is a generated projection and declares itself as one
generated, self-declaring, 97 KB, no placeholders
C07 PASS GOV-F2.3, GOV-G6.3, GOV-F4.2 Every relative link in the dashboard and Help Hub resolves on disk, and no absolute or drive-letter path is used
39 relative links, all resolve, none absolute
C08 PASS GOV-E5.1, INV-5, REQ-CON-06 No secret, credential, token or government identifier anywhere in the controlled set
zero secrets or identifiers across the controlled set
C09 PASS GOV-D4.3, GOV-D4.10 The study, the workbook, the dashboard and the registers report the same figures — checked by opening all four, not by matching strings in two
study, workbook, dashboard, README and registers agree on 12 checked figures
C10 PASS GOV-G3.4, REQ-SYS-14 The Help Hub troubleshooting section covers all nine failure modes named in GOV-G3.4
all nine GOV-G3.4 failure modes covered
C11 PASS GOV-G3.7, REQ-SYS-16, GOV-F2.2 Every dashboard metric has a lineage entry, and the published open-item count equals the count the registers actually produce
18 metrics with lineage in both surfaces; 12 dashboard tiles all map to a lineage row; open-item count 53 agrees everywhere
C12 PASS GOV-F5.1, REQ-SYS-13 Every diagram exists on disk and is embedded in the study
12 diagrams on disk, all distinct, all 12 embedded in the study by content hash
C13 PASS GOV-F3.7, GOV-F9.6 README.md is the single entry point and only it plus the dashboard sit at the project root
root holds only README.md and Dashboard.html; README carries all six required sections
C14 PASS GOV-F9.8 Every binary source of truth has a plain-text projection no older than it
27 csv projections, none staler than the workbook
C15 PASS GOV-F9.9 The handover file exists, is generated, and names the next action
handover file current, generated, names the next action
C16 PASS GOV-B5.4 Downward orphan detection: every requirement has a component and an evidence record
orphaned-down = 0 across 32 requirements
C17 PASS GOV-B5.5 Upward orphan detection: every evidence record and component points at a real requirement
orphaned-up = 0 across 32 evidence records
C18 PASS GOV-B4.2 Every component has exactly one responsibility, an owner and at least one requirement
13 components, each with one responsibility, an owner and requirements
C19 PASS GOV-B4.4, B4.5 Every interface has two endpoints, a direction, an item, and an owner on each side
22 interfaces, all with two endpoints and two owners
C20 PASS GOV-F4.2 No drive letter, absolute machine path, user name or machine name in any generated artefact
no machine-specific path in any generated artefact
C21 PASS GOV-F1.17, F1.18, F1.19 Every request carries the date, Zaid's verbatim words, the interpretation acted on, and the outcome
6 requests, each with all four mandatory fields
C22 PASS GOV-F9.14 Every file in the transfer pack re-derives from its source: nothing missing, extra or altered
15 pack files re-derive from their sources
C23 PASS GOV-F9.14, GOV-E1.6 NEGATIVE TEST — C22 is proven capable of failing when a pack file is altered
C22 detected a deliberate one-line tamper in HANDOVER.txt and the file was restored
C24 PASS GOV-C3.2 No artefact was verified by the agent that built it
8 artefacts, builder != verifier != validator on every one
C25 PASS GOV-C2.2 Every dispatched agent carries a charter with all six mandatory fields
8 agent charters, all six fields populated, no TBD
C26 PASS GOV-H2.5, GOV-C5.7 Every open item carries an owner and a due date or an explicit route
every open item has an owner and a route
C27 PASS GOV-F8.8, REQ-MOP-01 CLAIM ATTRIBUTION — every paragraph and table row of the study that states a dollar figure also cites a SRC or ASM identifier
attribution ratio 0.912 over 171 units carrying a dollar figure; 15 unattributed -> Budget of $15,000 to $30,000 to clear the startup phase [unsourced in the legacy files | CORE Capital to first revenue 4 $10,791 one-off, $15,875 six-month runway. | SC Capital to first revenue 5 $3,673 one-off, $8,757 six-month runway, no wage cycle. | A Capital at risk before first revenue 3 $17,117 six-month runway for the aged care enti | A Contribution per client per month 5 $1,002.06 per client per month against the NDIS en
C28 PASS GOV-F5.5, GOV-D4.10 The delivered workbook carries cached values, and its computed cost band equals the band the model computes
workbook carries 623 computed values; fixed-cost band (601.5, 847.33, 1438.17) matches the model exactly
C29 PASS GOV-F4.5a, REQ-CON-04 MOBILE LEGIBILITY over EVERY delivered Office artefact — the .docx, the .pptx and the .xlsx — not just the one that is easiest to open
4 delivered Office artefacts measured — NDIS_and_Aged_Care_Business_Enabling_Study_v2.0.docx (3980 runs, 50 tables); NDIS_and_Aged_Care_Decision_Pack_v2.0.pptx (148 runs, 12 slides); NDIS_and_Aged_Care_Financial_Model_v2.0.xlsx (1229 cells, 14 sheets); Project_Registers_v2.0.xlsx (5191 cells, 28 sheets) — zero runs or cells below 10pt in any of them, zero document tables above six columns. Widest WORKSHEET grid: Scenarios at 14 columns, and every worksheet above six columns freezes both a label column and a header row so scrolling does not lose them (REQ-CON-04 as amended under CR-007; DEC-008 superseded). Excluded as not this project's output: AI_Project_Governance_Standard_v3.7.docx, AI_Project_Governance_Standard_v3.9.md, AI_Project_Governance_Standard_v4.0_Business_Layer.md
C30 PASS GOV-F1.12, GOV-E1.4 Every evidence record points at an artefact that actually exists
32 evidence records, every cited artefact and sheet resolves on disk
C31 PASS GOV-E2.3, GOV-E2.4 NO SELF-CERTIFICATION — a requirement is Verified only where the INDEPENDENT verifier's recorded verdict says PASS
31 of 32 requirements independently verified; 1 recorded as still failing (REQ-CON-04); 0 AWAITING an independent verifier and correctly held at Open (none); verdict source: 02_Work/scratch/V10_closing_verification.md (independent verification pass 10, 2026-09-07, the closing pass against baseline v2.0.0. The progression across the five passes taken against this baseline was 23, then 30, then 30, then 30, then 31 of 32, and each pass found defects the one before it had missed)
C32 PASS GOV-D1.1, GOV-D1.2 Every configuration item points at a path that exists — including the Governance Standard itself
31 configuration items, every path resolves and no projection is stale
C33 PASS GOV-B4.10, REQ-BUS-001, REQ-BUS-009 The business plan twin exists, passes its own twelve checks, and the A3 page is not staler than the twin it projects
business plan twin passes 12 of 12 of its own checks; A3 page current; verdict 'DECISION OWED BY OWNER' consistent with verification state
C34 FAIL REQ-BUS-009, REQ-BUS-01 The business plan A3 page prints to EXACTLY ONE A3 landscape sheet — counted from a real PDF, not measured from a div
playwright is not installed, so the one-sheet claim cannot be printed and counted; an unmeasurable claim fails (GOV-C3.1)
C35 PASS GOV-F3.8, GOV-F3.9, GOV-F3.10, GOV-F3.11 File placement and debris — the tidy report is clean: no loose root files, no debris, no secret stores, no links to live files
placement map satisfied: 19 mandated folders present, root holds only Dashboard.html and README.md, no debris, no secret stores, no links to live files
C36 PASS GOV-D1.1, GOV-F8.8 Every SRC and ASM identifier cited by a financial model resolves to a row that actually exists in the registers
55 distinct source and assumption identifiers cited across the models, every one resolves against 105 register rows
C37 PASS GOV-B4.1, GOV-D4.10 The workbook's aged care sheet and the aged care model agree, figure by figure, on the numbers the recommendation rests on
15 aged care and NDIS per-client figures agree between the workbook and the model to the cent
C38 PASS GOV-B4.10, GOV-B4.9 The system breakdown exists, is current, and every component in it carries at least one requirement that actually exists
system breakdown current: 13 components, every one with a requirement that exists; 22 interfaces mapped
C39 PASS GOV-D4.11 Every generated document and surface names its generator (OWNER) and the trigger that regenerates it (REVIEW)
8 generated documents and surfaces, every one naming its generator and its regeneration trigger; CSV projections exempt with a stated reason
C40 PASS GOV-D1.13 A source manifest exists and every source of truth matches the hash recorded at the last gate
20 sources of truth all match the manifest gated at 2026-09-07T09:18:25
C41 PASS GOV-E2.3, GOV-E2.4, GOV-D5.2 No requirement is marked Verified on a verdict taken against a DIFFERENT baseline from the one the project is currently at
32 of 32 requirements carry a verdict taken against the CURRENT baseline v2.0.0; 0 carry a verdict retired with an earlier baseline and are reported as stale rather than as passing (none)
C42 PASS GOV-D4.11, GOV-B4.1 Every register projection matches a FRESH projection of the source of truth, and no generated artefact predates the source it projects
619 register rows across 27 projections match a fresh projection of the source of truth exactly, and all 53 generated artefacts were written after the newest of 14 sources
C43 PASS GOV-B6.1, GOV-B6.3, REQ-SYS-03, REQ-SYS-04 Every trade study total published in the delivered study is recomputed from its own scores and weights, and the weights sum to one
10 trade study totals across 3 studies recomputed from their own scores and weights and found in the delivered study; every score carries the figure it rests on; every weight set sums to 1.0
C44 PASS GOV-D4.3, GOV-F8.8 No generated surface carries an unresolved template token, and every count it publishes about an artefact matches that artefact
no unresolved template tokens across 5 generated surfaces; the published counts match the artefacts — 12 diagrams, 12 slides, 455 workbook formulas, 83 study pages
FAILURES: 1 of 44 checks failed