Thirty tagged claims — what the campaign thought it knew, how sure it was, and what would break if it was wrong. The most transferable thing the project built.

Document begins

Bottom Pub Co-op — Assumptions Register

Internal working document. Not for public release. Gitignored. Every entry uses the decision-discipline categories from AGENTS.md: FACT, ASSUMPTION, CHOICE, RISK, TODO. This register was seeded from the multi-agent review and a manual scan of docs/website_content/, docs/proposals/, and internal/workplans.md. It should be updated each time a new claim enters any project document.

How to read this

Column	Meaning
ID	Stable identifier. Once assigned, does not change.
Category	FACT (verified from source) / ASSUMPTION (plausible, unverified) / CHOICE (strategic preference, needs decision) / RISK (could harm credibility or legal position) / TODO (needs professional advice or missing data)
Claim	The specific assertion as it appears or is implied in project materials
Source	Where the claim came from: meeting transcript, website, internal discussion, briefing, public record, or unverified
Confidence note	How confident we are and what kind of verification would shift the category
If wrong	What changes for the project if this claim turns out to be incorrect

Seeded entries (multi-agent review)

ID	Category	Claim	Source	Confidence note	If wrong
AR-01	ASSUMPTION	The Commercial Hotel heritage listing is at state level (Tasmanian Heritage Register), not national level (National Heritage List or Commonwealth Heritage List).	Meeting transcript: builder’s verbal statement (“The heritage facade is at state level, not national level as far as I know”). Website (index.html) reports it as meeting-reported and not independently verified, citing only the confirmed THR Place ID 3472.	Plausible but verbal from a builder, not a heritage professional. Verified fact: building is on THR (Place ID 3472). Not verified: that it is NOT on any national list. Cross-check with the Australian Heritage Database at environment.gov.au.	If the building is also on a national list, the approval pathway changes — Commonwealth heritage provisions under the EPBC Act would apply in addition to the Tasmanian Heritage Council process. This could add federal referral steps, longer timelines, and different significance tests.
AR-02	ASSUMPTION	Earlier offers in the range of approximately $1.6 million were reportedly rejected for the pub without the residential land.	Website (index.html) says “reportedly rejected” but the qualifying context — that the $1.6M was for the pub alone, not including the ~0.5 ha residential land — is not prominent on the website. Meeting transcript includes the fuller context.	Meeting-reported, third-hand. No formal valuation or agent statement has been seen. The $1.6M figure could be misread as a likely co-op entry price if the residential-land qualifier is lost.	If the $1.6M figure is wrong or no longer relevant (e.g., the insolvency process changes the asking price), any capital model built on it is unreliable. If the qualifier is lost, community members may assume the entry price is ~$1.6M when the actual all-in cost (pub + land) could be materially different.
AR-03	ASSUMPTION	Membership or share contributions might be pitched at around $200–$500 per member.	Meeting transcript: “if someone put in $200 to $500, that could make them a member.” Website (index.html) reports this as “a separate hypothetical example discussed in the room” and notes it is from “a different town, a different building, and a different market.” The meeting attributed this range to the Cygnet discussion, not to Castlemaine’s actual pricing.	Local hypothetical for Cygnet, not Castlemaine’s actual membership pricing. The index.html text now separates this from the Castlemaine debenture figures, but the meeting transcript shows the two were discussed adjacently, creating risk of conflation. The Castlemaine co-op (Castlemaine Community Investment Co-operative Ltd) does not sell memberships in the same sense — it issues debentures at $1.95M–$2.0M total (verified). Membership/share pricing for that co-op has not been extracted from primary sources.	If quoted as Castlemaine’s price on any document, it is a misattribution that undermines case-study credibility. If used to model Cygnet’s capital raise, $200–$500/member is a design choice, not evidence — it needs its own feasibility test.
AR-04	ASSUMPTION	Around 81 cents in the dollar stays locally in a co-operative model, versus a much smaller amount under external ownership.	Meeting transcript: “A speaker cited a comparison of around 81 cents in the dollar staying locally in a cooperative model versus a much smaller amount under external ownership.” Website audit page: “No primary source for the specific ‘81 cents in the dollar’ figure has been located.” Not cited on the public site outside the audit page.	No primary source found. The NEF Local Multiplier 3 (LM3) methodology is the most plausible origin, but no document citing that precise figure for a co-operative model has been located. Related LM3 research cites different figures (67–68 cents at independent local businesses vs ~43 cents at chains).	The statistic is a powerful community-ownership argument. If it is wrong or cannot be sourced, the project loses a key rhetorical claim. If a primary source is later found, it can be promoted to FACT; if not, it must not be used in any public-facing material.
AR-05	FACT	The Commercial Hotel is publicly listed for sale by EOI through Elders Real Estate Hobart, on behalf of the mortgagee in possession. EOI closes 2 July 2026 at 16:00.	commercialrealestate.com.au, property ID 2020889392, last updated 3 June 2026. Listing title: “Historic Hotel with Substantial Landholding, Full Repositioning Potential.” 5,096sqm* freehold, 738sqm* improvements, 2,400sqm* described as residential-zoned surplus land. Updated 12 June 2026: confirmed from live listing. Prior status (18 May 2026) was unconfirmed.	FACT — confirmed from public listing. The relationship between Elders (appointed agent) and [firm name withheld] (previously identified insolvency practitioner) has not been independently confirmed from the listing alone.	EOI deadline of 2 July 2026 is the single most time-critical dependency. If the co-op does not submit an EOI or engage Elders before that date, it may lose standing in the formal process. The property could sell to another buyer; the steering committee must act before close.

Entries surfaced from document scan

The following entries were identified by scanning docs/website_content/, docs/proposals/, and internal/workplans.md for claims that lack explicit source labels or are stated without attribution to a decision-discipline category.

ID	Category	Claim	Source	Confidence note	If wrong
AR-06	ASSUMPTION (downgraded 2026-05-22 — pending verification)	The parcel-specific zone for the Commercial Hotel site has not been verified against the operative Huon Valley LPS zoning map on PlanBuild Tasmania. Earlier drafts assumed General Business Zone; the geospatial briefing records overlays in the parcel’s vicinity that include Local Business, General Residential, Environmental Management, Future Urban, and Utilities.	docs/website_content/planning_and_development_briefing.md (revised 2026-05-22) now flags the zone as “[Zone overlay still being confirmed against PlanBuild Tasmania]” and conflicts with docs/website_content/geospatial_briefing.md overlay inventory.	Conflicting briefings; not yet reconciled against operative LPS. Public copy now reflects this uncertainty rather than asserting General Business Zone.	If the zone is different from General Business, the use status, required permits, and applicable codes all change. Entertainment venue may not be discretionary, or food/drink premises may not be permitted — both of which alter the feasibility pathway. Reconciliation against PlanBuild Tasmania is a Stage 1 / Stage 2 boundary task.
AR-07	ASSUMPTION	The commercial building and ~0.5 ha residential land are on a single title that “may be handled as a single title” — not yet confirmed.	internal/workplans.md (task 1.4): “Reported at the meeting as ‘may be handled as a single title’ — not yet confirmed. The title search will answer this definitively.”	Meeting-reported, unverified. Title search not yet conducted.	If the title can be split, the co-op might acquire the pub alone at lower cost and the residential land could be developed or retained separately. If it cannot be split, the purchase price includes the residential land, changing the capital requirement.
AR-08	ASSUMPTION	Renovation costs for the Commercial Hotel are in the range of $900K–$1.2M.	internal/workplans.md (Stream 3, task 3.1): “Renovation range ($900K–$1.2M, meeting-reported, not verified).” No independent building assessment or quantity surveyor estimate supports this.	Meeting-reported. Third-hand. No independent assessment.	If renovation costs are materially higher (common for heritage buildings with BCA change-of-use upgrades), the capital model breaks. The feasibility briefing warns that heritage upgrades routinely exceed non-heritage commercial estimates.
AR-09	ASSUMPTION	A previous buyer ([name withheld — private individual]) paid approximately $1.3M.	internal/workplans.md (Stream 3, task 3.1): “Purchase price range ($1.3M previous buyer, $1.6M earlier offers reported — meeting-reported, not verified).”	Meeting-reported. Not confirmed from title records or sale documents.	If the actual purchase price was different, the $1.3M figure misleads capital modelling.
AR-10	ASSUMPTION	Lachlan’s verbal building assessment (kitchen workable but not ideal, upstairs needs total renewal, asbestos issues, veranda $100K+, freezers/refrigeration need upgrades, outbuildings degrading, heritage facade constraints) is an accurate representation of the building condition.	internal/workplans.md (Stream 4, task 4.1): described as “Lachlan’s verbal assessment from the meeting” with guardrail: “Don’t treat Lachlan’s verbal assessment or [the previous prospective purchaser]‘s documents as a formal condition report. They’re real data from informed people, but they’re not a commissioned independent assessment. Label them as meeting-reported and third-party respectively.”	Meeting-reported verbal assessment from a local builder with 26 years in Cygnet, who had inspected the building. Not a commissioned report. No independent verification.	If the building condition is materially worse (or better) than described, the renovation scope and cost estimates change. A heritage-listed building with unknown structural condition is the key feasibility unknown.
AR-11	ASSUMPTION	The veranda repair cost is approximately $100K+.	Meeting transcript via internal/workplans.md (Stream 4, task 4.1 notes). Lachlan’s verbal estimate.	Single verbal estimate, no quantity surveyor or heritage architect input. Not adjusted for heritage-fabric compliance (veranda is likely significant fabric requiring Heritage Council approval).	Heritage-compliant veranda work could cost significantly more than $100K if the Heritage Council requires like-for-like materials or specific methods. This figure is a lower bound, not a budget line.
AR-12	CORRECTED	The meeting transcript reported that “the Castlemaine co-operative raised approximately $1.5 million through community debentures.” The verified public figure is different.	Verified primary sources (2026-05-18): Entity = Castlemaine Community Investment Co-operative Ltd (ABN verified via ABN Lookup). Asset = The Hub, 233 Barker St, Castlemaine (19 office spaces, meeting room, kitchen, cafe, community garden — NOT a hotel or pub). Co-op’s own disclosure statement: debenture raise minimum $1.95M, maximum $2.0M. ABC reporting confirms the $1.95M target. Meeting transcript’s “$1.5 million” figure is unverified and inconsistent with the public offer documents.	The meeting transcript’s $1.5M figure is lower than the co-op’s actual public offer ($1.95M–$2.0M). Additionally, the asset is not a “local hotel” — it is a mixed-use commercial building (The Hub). Both discrepancies must be corrected wherever the meeting figure has been used as a capital benchmark.	Do not use $1.5M as the Castlemaine debenture precedent. The verified benchmark is $1.95M–$2.0M from the co-op’s disclosure statement. The asset is also non-analogous to a pub: The Hub is an office/community space, not a hospitality venue.
AR-13	CORRECTED	The meeting transcript reported “optional interest of up to around 4–5% over 10–15 years” for Castlemaine debentures. The verified public terms are different.	Verified primary source (2026-05-18): Castlemaine Community Investment Co-operative Ltd disclosure statement lists interest options of 0%, 1%, 2%, 3%, or 4% annually, with 5/10/15-year terms. The maximum is 4%, not 5%.	The meeting-reported “up to 5%” figure is one percentage point above the co-op’s actual maximum offer of 4%. This is a small but material discrepancy for debenture design: if Cygnet models a 5% ceiling based on this precedent, it is citing a figure that the actual co-op did not offer.	Use the verified range (0–4%, 5/10/15-year terms) as the Castlemaine debenture benchmark. Do not cite “up to 5%” as a precedent. The website’s existing framing (“meeting context, not an offer”) remains correct; the corrected figure is now available from the primary source.
AR-14	CHOICE	The co-operative structure will be either distributing or non-distributing. This is the single most consequential legal decision and has not been made.	Stated across governance briefing, legal briefing, financial briefing, capital pathways briefing, and workplans.md. Not a claim to verify — it is an unmade decision that gates everything downstream.	Not yet a claim — it is a pending decision. The risk is that either option is assumed rather than decided. The audit page records that an earlier draft attempted to straddle both (5% dividends + community grants + surplus reinvestment), which is CNL-non-compliant.	If the project proceeds without making this choice explicitly and on legal advice, the rules, fundraising language, and member expectations may be legally incompatible. Irreversible without restructuring.
AR-15	ASSUMPTION	A co-operative of this scale in Cygnet would likely qualify as a base-rate entity (25% corporate tax rate) under ITAA 1997.	docs/website_content/accounting_and_tax_briefing.md: “A community pub co-operative generating predominantly active hospitality trading revenue … should easily satisfy the passive income test.” The briefing is careful to state “should easily satisfy” rather than “will satisfy,” and eligibility is assessed annually.	Plausible for a single-site pub under $50M turnover. But eligibility must be annually confirmed and depends on the actual income mix (rent from any tenanted portions, investment income).	If the co-op’s passive income exceeds 20% in any year (e.g., significant rental income from a tenanted model), the 30% rate applies, increasing the tax burden by 5 percentage points on assessable income.
AR-16	ASSUMPTION	Approximate $1.5–$2.5M is a reasonable indicative property value for bridging-finance modelling.	docs/website_content/social_finance_briefing.md (section 7): “For a property worth approximately $1.5–2.5 million (ASSUMPTION — no current valuation exists)…” The briefing explicitly labels this as ASSUMPTION and notes no current valuation exists.	Explicitly unvalued. Range appears to be bracketed from the meeting-reported $1.3M and $1.6M figures plus an upward estimate that includes the residential land.	If the actual market value is outside this range (likely, given no independent valuation), bridging-finance cost estimates and LVR calculations are wrong. The $58,500–$97,500 interest-cost range in the social finance briefing rests on this assumption.
AR-17	ASSUMPTION	Food premises registration fees for a higher-risk commercial kitchen at Huon Valley Council are typically in the range of $300–$700 per year.	docs/website_content/planning_and_development_briefing.md: “ASSUMPTION. Based on comparable Tasmanian councils… Retrieve the current Huon Valley Council Fees and Charges schedule directly for the exact figure.”	Based on comparable-council benchmarking, not HVC’s actual schedule. Small dollar figure, but the principle matters: every cost line in the model should be sourced.	Wrong direction for feasibility: if the actual fee is higher, the operating-budget error is minor in absolute terms but the pattern of unsourced cost lines is concerning. If lower, the model is slightly conservative, which is acceptable.
AR-18	ASSUMPTION	Development application fees for a commercial change-of-use at HVC are likely in the range of $500–$2,500 depending on estimated project value.	docs/website_content/planning_and_development_briefing.md: “ASSUMPTION: fees for a commercial property of this scale are likely in the range of $500–$2,500 depending on estimated project value. Retrieve the current schedule directly before any feasibility model is finalised.”	Same as AR-17: comparable-council benchmarking, not the actual HVC schedule.	If the actual fee schedule is different (especially if it is a percentage-of-project-value model common in some councils), the fee could be materially higher for a $1M+ renovation project.
AR-19	ASSUMPTION	The Cygnet community is larger and the Huon Valley is wealthier on a per-capita basis than Grong Grong (population ~100), making the Grong Grong capital-raise precedent ($1M+ from 169 members) a plausible benchmark.	docs/website_content/social_finance_briefing.md (section 8): “The Cygnet community is larger and the Huon Valley is wealthier on a per-capita basis than Grong Grong (population ~100).”	Plausible qualitative comparison. No per-capita income or wealth data is cited. “Larger” and “wealthier” are reasonable inferences from census and regional data but not demonstrated in the document.	If the comparison is wrong or the wealth distribution in Cygnet is different (e.g., more retirees on fixed income, fewer high-net-worth individuals), the $1M+ member-capital benchmark may not be realistic. The social finance briefing uses Grong Grong as “the best Australian model” and “demonstrably replicable” — that claim rests on this comparison.
AR-20	ASSUMPTION	Heritage project approval (pre-application to permit grant) typically takes 3–6 months for complex adaptive-reuse projects.	docs/website_content/planning_and_development_briefing.md (section 4): “For complex adaptive reuse projects, budget 3–6 months from pre-application to permit grant.” Based on Heritage Tasmania guidelines and the statutory clock (up to 56 days for DA + Heritage Council assessment).	Based on statutory timeframes and Heritage Tasmania’s own guidance on complex projects. Reasonable as an indicative planning estimate.	If the Heritage Council requires a Conservation Management Plan, additional information requests, or objects to proposed works, the timeline extends. Heritage refusal is a hard stop (Council cannot grant a permit against Heritage Council refusal).
AR-21	ASSUMPTION	The acoustics management plan (AMP) for the DA will cost approximately $3,000–$8,000 depending on complexity.	docs/website_content/planning_and_development_briefing.md (section 6): “Budget approximately $3,000–$8,000 for an AMP report depending on complexity.”	No quote obtained. Based on industry rates for NATA-accredited acoustic consultants on comparable projects.	If the actual cost is higher (especially if a complex heritage building requires extensive measurement or modelling), this feasibility line item is low.
AR-22	RISK	An earlier internal draft proposed paying community investors a 5% annual dividend, returning capital at Year 15, and securing the offer with a first mortgage over the property. These claims have been removed from all public-facing documents, but the audit page records them, and any community members who saw the earlier draft may still hold these expectations.	docs/website_content/financial_and_asic_briefing.md (section 1); docs/website_content/governance_contradictions.csv; site/audit.html.	The claims are confirmed as removed from public material. The risk is residual community expectation, not current publication.	If community members enter a future capital raise expecting the terms from the earlier draft, the co-op faces a trust problem even if the legal documents are different. The audit page and meeting record are the mitigation — they acknowledge the change publicly.
AR-23	RISK	The sociocratic governance model (circles, double-linking, consent-based decisions) from the earlier proposal is incompatible with CNL statutory requirements (special resolutions require 75% majority regardless of internal consent models).	docs/website_content/governance_briefing.md (section 4); docs/website_content/governance_contradictions.csv row on “Consent-based decision making (Sociocracy)”.	Confirmed as a legal risk by the governance briefing and the contradictions CSV. The governance briefing recommends a standard elected-board model. Not a claim to verify — a flagged risk.	If sociocratic elements are re-introduced in the rules without reconciling them to CNL voting requirements, the rules may not be approved by the Registrar. More broadly, the governance choice must be explicit and documented.
AR-24	RISK	Hotel Theodore (Qld) — the first confirmed Australian community-pub failure (entered voluntary administration August 2023 after ~75 years) — demonstrates that community ownership does not remove hospitality cost-structure risk.	docs/website_content/australian_community_pub_case_studies.md; docs/website_content/capital_pathways_briefing.md; docs/website_content/social_finance_briefing.md. All three cite BCCM Reinvigorating Rural Australia as source.	Verified from BCCM source. Not a claim to verify — it is a precedent that must be kept in the public record. The risk is that it is downplayed or omitted in future community materials.	If the project does not stress-test wages, energy, insurance, repairs, and management depth (the factors BCCM identifies), it repeats the same pattern. This failure case is essential honest evidence.
AR-25	TODO	The exact protected fabric of Tasmanian Heritage Register Place ID 3472 has not been quoted verbatim in the project briefings. The heritage briefing paraphrases rather than quotes the register entry.	docs/website_content/heritage_and_building_briefing.md (section 5): “The exact protected fabric of Place ID 3472… This briefing still paraphrases the protected fabric rather than quoting it verbatim.”	The register entry and datasheet are linked. The paraphrase is conservative. But for Heritage Council applications, the exact wording of the statement of significance matters.	If the paraphrase omits or misstates a protected element, the feasibility budget may not include work that the Heritage Council will require, or may include work that the Heritage Council will refuse.
AR-26	TODO	No independent building-condition assessment has been commissioned. Without it, no realistic capex estimate is possible.	docs/website_content/heritage_and_building_briefing.md (section 5); internal/workplans.md (Stream 4, tasks 4.4, 4.6). Both flag this as the critical next step.	Stated as unknown. This is the single largest source of capex uncertainty.	Any capex figure cited without an independent assessment is a meeting-reported estimate, not a budget. The renovation range ($900K–$1.2M, AR-08) cannot be confirmed without this.
AR-27	TODO	The current status of Tasmania’s liquor licensing reform (announced March 2025, proposing a new entertainment endorsement) needs confirmation before advising on licensing structure.	docs/website_content/planning_and_development_briefing.md (section 6): “As of May 2026, the reform legislation was still before Parliament — confirm current enactment status.”	Legislative status verified as at briefing date. Bills can pass, lapse, or be amended.	If the reform has passed or been substantially amended, the licensing pathway (entertainment endorsement, fee structure, application process) may differ from what the briefing describes.
AR-28	TODO	Whether a community trading pub can achieve and maintain ACNC charitable registration is a high-risk pathway requiring specialist charity law advice.	docs/website_content/accounting_and_tax_briefing.md (section 3): flagged as [PROFESSIONAL ADVICE REQUIRED]. The briefing notes the ACNC requires all purposes to be charitable, and a commercial pub trading activity may not straightforwardly satisfy this test.	Not verified. The briefing explicitly recommends specialist advice before any constitution is drafted on this basis.	If the project assumes charitable status for tax modelling and it is later refused by ACNC, the tax position changes materially (no income tax exemption, no FBT concessions, no DGR status). The structure choice (distributing vs non-distributing) must be made without assuming charitable status unless confirmed.
AR-29	ASSUMPTION	Approximately 68 EOIs and ~12 steering committee volunteers were recorded at the time of the 17 May 2026 meeting.	Website (index.html): “approximately 68 expressions of interest had already been received at the time of the meeting, including around 12 people offering to join the steering committee.” Meeting transcript and internal/workplans.md confirm same numbers.	Meeting-day count from the steering group’s EOI form. Verifiable from the EOI database (held by Billie). Not independently audited by the project.	If the EOI database has duplicate entries, invalid contacts, or significant drop-offs since meeting day, the 68 figure overstates current support. The number should be updated (workplans task 5.1) and not treated as a static fact.
AR-30	ASSUMPTION	Bank Australia’s commercial variable rates for commercial property lending are broadly in the 7–9% p.a. range.	docs/website_content/social_finance_briefing.md (section 1): “APPROX — exact rates are not publicly advertised and must be sought by application.”	Explicitly labelled as approximate. Based on market rates post-RBA 2025–26 movements. Not a quote from Bank Australia.	Actual rate at application may differ significantly. If rates have moved or the co-op’s risk profile pushes the rate above 9%, debt service costs increase and the capital model may not hold.

Cross-references

Entry	Related entries	Where it appears in project docs
AR-01 (heritage listing level)	AR-20, AR-25, AR-26	index.html, heritage_and_building_briefing.md, meeting-17-may.html
AR-02 ($1.6M figure)	AR-07, AR-09, AR-16	index.html, workplans.md (task 3.1), social_finance_briefing.md
AR-03 ($200–$500 membership)	AR-12, AR-13	index.html, meeting-17-may.html
AR-04 (81 cents in the dollar)	—	meeting-17-may.html, audit.html
AR-05 (property availability)	AR-02, AR-07, AR-09	index.html, workplans.md (Stream 1), proposals/public_eoi_draft_v1.md
AR-06 (zoning)	AR-20	planning_and_development_briefing.md
AR-07 (single title)	AR-02, AR-05	workplans.md (task 1.4)
AR-08 (renovation range)	AR-10, AR-11, AR-26	workplans.md (task 3.1)
AR-10 (Lachlan’s assessment)	AR-08, AR-11, AR-26	workplans.md (Stream 4, task 4.1)
AR-14 (structure choice)	AR-22, AR-23, AR-28	legal_and_licensing_briefing.md, governance_briefing.md, financial_and_asic_briefing.md, workplans.md (Stream 2)
AR-16 (property value range)	AR-02, AR-09	social_finance_briefing.md (section 7)
AR-22 (removed financial claims)	AR-14, AR-23	financial_and_asic_briefing.md, governance_contradictions.csv, audit.html
AR-26 (no building assessment)	AR-08, AR-10, AR-11	heritage_and_building_briefing.md, workplans.md (Stream 4)
AR-28 (charitable status)	AR-14, AR-15	accounting_and_tax_briefing.md

Update protocol

When a new claim enters any project document (briefing, website, meeting record, workplan), add it to this register with a category, source, confidence note, and impact-of-being-wrong.
When a claim is verified from a primary source, promote it: ASSUMPTION → FACT. Record the source and date.
When a claim is disproved, change its category to RISK and note the correction.
When a TODO is resolved through professional advice, update the entry with the advice outcome and the adviser’s name/date.
Do not remove entries — mark them superseded if they are replaced.
This register is internal-only. Do not link to it from any public-facing page or commit it to an un-ignored path.