@rubytech/create-sitedesk-code 0.1.340

package/payload/premium-plugins/venture-studio/skills/investor-data-room/SKILL.md

@@ -0,0 +1,390 @@
+---
+name: investor-data-room
+description: Use when a UK-domiciled founder needs to produce a full seed-raise pack from idea to deliverables: graph-structured data room, 16-section business plan, persuasive prospectus + formal term sheet (with FCA self-certification appendices), 13-slide deck blueprint, print-ready HTML+PDF for each. Trigger phrases include "build me a data room", "draft a business plan", "deck from the business plan", "draft an investment prospectus", "generate a term sheet", "from office hours to investor pack", "set up the fundraise repo". The skill never speculates on financial figures or cap-table values: every number traces to a Companies House filing, a financial model worksheet, or a source citation. Public-facing artefacts contain zero internal workings (no "reconciliation pending", no "open Q" markers, no comparison-to-baseline commentary). Design tokens are harmonised across business plan, prospectus, term sheet, and deck via the operator's project design-tokens spec.
+allowed-tools:
+  - Bash
+  - Read
+  - Write
+  - Edit
+  - Glob
+  - Grep
+  - WebSearch
+  - WebFetch
+---
+
+# investor-data-room
+
+Build a UK seed-raise pack in five stages. Each stage has an outcome contract; do not advance until the contract is met.
+
+## Outcome contract (whole skill)
+
+Complete when the data room contains:
+
+```
+<project-site>/.docs/data-room/
+  README.md                                 # data-room index, status snapshot, doctrine
+  01-narrative/
+    office-hours-design.md                  # ideation working scratchpad
+    business-plan.md                        # 16-section business plan
+    deck-blueprint.md                       # 13-slide blueprint
+    exit-thesis.md                          # named acquirers + recent exit comparables
+  02-corporate-legal/
+    README.md                               # incorporation filings, articles, board minutes
+    incorporation-<co-house-no>-<date>.pdf  # IN01 + statement of capital + cap table
+  03-cap-table/
+    README.md                               # pre/post-investment cap table with PSC notes
+    comparables.md                          # recent comparable rounds (company / pre-money / post-money / multiple / source) cited by the term-sheet valuation
+  04-financials/
+    README.md
+    model_worksheet.md                      # opening cash, capex, burn, MRR ramp, two-round
+  05-commercial/
+    README.md
+    <gtm-partner-1>.md                      # one file per named partner
+    <gtm-partner-1>-proposal.pdf
+  06-product-ip/
+    README.md
+    encrypted-document-sharing.md           # if applicable
+  07-team/
+    README.md
+    <founder-1>.pdf                         # CV per foregrounded founder
+  08-market/
+    README.md
+    sources.md                              # every market figure with citation
+    competitor-value-claims-rebuttal.md
+  09-operations/
+    README.md
+  10-supporting/
+    README.md
+  html/
+    business-plan/
+      index.html                            # A4 portrait, design tokens, premium register
+      render-pdf.mjs                        # Playwright + img2pdf + qpdf pipeline
+      business-plan.pdf                     # rendered output
+      logo-light.png logo-dark.png
+    prospectus/
+      index.html                            # A4 portrait, 10-15 page persuasive cut
+      term_sheet.html                       # A4 portrait, 6-8 page formal term sheet
+      render-pdf.mjs
+      prospectus.pdf term_sheet.pdf
+      logo-light.png logo-dark.png
+    deck/
+      index.html                            # 16:9 landscape (297mm x 167mm), 12+1 slides
+      render-pdf.mjs
+      deck.pdf
+      logo-light.png logo-dark.png
+      team/                                 # cropped headshot WebPs for slide 8 founder grid
+        <founder-firstname>-<lastname>.webp
+```
+
+Skip stages that are already complete; do not regenerate stale artefacts blindly.
+
+## Stage 1 — Office-hours ideation
+
+Drive a conversational discovery session using the sibling `office-hours` skill. Market sizing and competitor recon during discovery dispatch the `research-assistant` specialist with `deep-research` + `data-research` rather than calling `WebSearch` directly — every market figure that ends up in `08-market/sources.md` must trace to a structured-research citation.
+
+Output a single `office-hours-design.md` capturing:
+
+- Problem statement
+- Demand evidence (with numbers and dates)
+- Status quo
+- Target user and narrowest wedge
+- Constraints
+- Six to eight numbered premises (founder-confirmed)
+- Two to four narrative-shape alternatives with a recommendation
+- Financials / use-of-funds placeholders flagged as open questions
+- Founder Signal Reflection
+
+This file is the working substrate. It stays internal-only. Every later artefact derives from it.
+
+**Doctrine.** Never speculate on numbers; flag them as open questions until the founder confirms.
+
+## Stage 2 — Data room scaffold
+
+When invoked under the venture-studio agent, scaffolding is done by the deterministic script at `premium-plugins/venture-studio/bin/scaffold.sh` — it creates the Project + eight artefact Tasks and the ten-section directory tree in one atomic invocation. The script is idempotent; re-runs against an existing data room are a no-op. See `venture-studio/PLUGIN.md` § "Scaffolding script".
+
+When invoked outside the venture-studio agent (standalone usage of this skill), build the directory tree manually. Each section gets a `README.md` that doubles as a content checklist (purpose, expected contents, current state, gaps). The graph structure is fixed: ten numbered sections (`01-narrative` through `10-supporting`) plus a top-level README and an `html/` subdirectory for rendered deliverables.
+
+**File-naming conventions:**
+
+- `<co-house-no>` for Companies House filings (e.g. `incorporation-17237239-2026-05-22.pdf`).
+- `<founder-firstname>-<lastname>.pdf` for CVs.
+- `<gtm-partner>.md` + `<gtm-partner>-proposal.pdf` for each commercial partner.
+- All `README.md` files are markdown only.
+
+**Where the published URL lives.** If the project hosts a public prospectus or marketing microsite, keep the published artefacts at a stable path (e.g. `realagent.chat/sites/<slug>/`); the rendered HTML inside `html/<artefact>/index.html` is the source of truth and republishes from there.
+
+References: `references/data-room-structure.md` for the section-by-section purpose statements.
+
+## Stage 3 — Business plan synthesis
+
+**Before drafting any narrative prose**, call `voice-retrieve-conditioning` with `format: "article"`, `length: "long"`, `scope: "org"`, and `topic` set to the venture's domain (e.g. "AI estate-agent operating system"). An investor document goes out under the venture's name, so it uses the **house voice** (`scope: "org"`), not any one founder's personal voice. When the tool returns a non-null `styleCard` (`status` of `ok` or `fallback-org`), condition every prose section on the style card and exemplars so the business plan reads as the house's own writing, not generic VC boilerplate. When `status` is `no-data` or `error`, draft in the formal-yet-direct register implied by the section purpose statements without further conditioning. Emit narration `[investor-data-room] voice-mirror styleCardBytes=<n> exemplars=<k>` when conditioning was applied.
+
+Generate `01-narrative/business-plan.md` with **16 sections** following the canonical structure:
+
+1. Executive Summary
+2. Vision and Mission (Vision / Mission / Doctrine triad)
+3. The Company (legal entity table + pre-investment cap table)
+4. The Market (TAM/SAM/SOM bottoms-up with geography-tier floor + CAGR with citation + 3-yr market-expansion thesis + market-share trajectory to 3–8% SAM in 5 yrs + structural tailwind + self-employed wedge + domestic adjacency + international upside + pricing)
+5. The Product (what it is + wedge product + upsell path + substrate moat + conversation ingestion + network + Anthropic asymmetric capture + human-services layer)
+6. The Business Model (revenue streams + unit economics — ACV, CAC, LTV, **LTV:CAC**, gross margin, **burn multiple**, churn — each a named line)
+7. Go-to-Market (Wedge → Beachhead → Blitzscale framework + demand evidence + GTM partners + horse-before-cart logic + geography)
+8. Competition (three failing categories + RANL differentiation + most-direct-competitor counter)
+9. Operations (technology stack + data sovereignty + IP ownership chain + compliance posture summary)
+10. Compliance and regulatory positioning (UK consumer protection + UK data/AI regime + EU AI Act + adjacent UK obligations + moat-not-tax close)
+11. Team (founders + strategic shareholders + hiring plan)
+12. Financial Plan (12.1 opening balance sheet through 12.7 use of funds; valuation per round cites `03-cap-table/comparables.md`)
+13. Risks and Mitigations
+14. Milestones (the N-month thesis)
+14b. Exit thesis (target exit value + 5–8-year horizon + exit routes — strategic / PE / IPO — + 3–5 named acquirers with one-line rationale each)
+15. The bottom line (summary table + close)
++ Appendices (titled documents only; no internal-only file pointers)
+
+**Critical rules** (all enforced; recurrence is a P0 violation):
+
+- **No internal workings** in body text. No "reconciliation pending", no "open Q", no comparison-to-baseline ("loss has grown from X to Y because…"), no cross-references to `office-hours-design.md` or any working file.
+- **Cap table is authoritative from the IN01.** Reconcile the share-capital line (`500,000 ordinary @ £0.001 nominal` vs `@ £0.01`) before writing the opening balance sheet. Inception premium (e.g. one founder paying £4,500 above nominal) reconciles £5k inception cash against £500 nominal.
+- **Every market figure cites a source.** Maintain `08-market/sources.md` with every figure used in the business plan; the body references "Appendix B" by title, never an internal file path.
+- **Financial figures derive from the model worksheet.** Cite "Appendix A" by title.
+- **No em-dashes between alphabetic words** in user-facing prose (recurring style violation in this project's history; see `references/internal-workings-scrub.md`).
+
+**Research routing across Stage 3.** Every figure sourced from the wider world routes through the `research-assistant` specialist — never raw `WebSearch` / `WebFetch` from this skill:
+
+- **Section 4 (Market)** — TAM/SAM/SOM citations, CAGR sourcing, market-share trajectory baselines, competitor rebuttal facts: dispatch `research-assistant` with `data-research`.
+- **Section 10 (Compliance)** — DMCC Act (UK consumer protection for estate-agency), ICO ADM framework (Articles 22A–22D under UK GDPR + ICO 2025 strategy), EU AI Act (force date, high-risk classification for property-financing decisions): dispatch `research-assistant` with `academic-verify` to retrieve the primary regulatory texts. Cite each in the body.
+- **Section 12 (Financial — valuation comparables)** — recent comparable rounds that feed `03-cap-table/comparables.md`: dispatch `research-assistant` with `data-research`.
+- **Section 14b (Exit thesis)** — 3–5 named acquirers (strategic / PE / IPO) with one-line rationale each, plus recent comparable exits (company / acquirer / multiple / year / source) that feed the exit section of `03-cap-table/comparables.md`: dispatch `research-assistant` with `data-research`.
+
+The specialist returns structured findings; the skill quotes those findings verbatim into the relevant section and into `08-market/sources.md` / `03-cap-table/comparables.md`.
+
+References: `references/business-plan-template.md` for the section-by-section template; `references/compliance-research-checklist.md` for the UK + EU regulatory regimes to verify.
+
+## Stage 3b — Term sheet synthesis
+
+Generate `html/prospectus/term_sheet.html` as an **A4 portrait, 10-page formal term sheet** sectioned as:
+
+1. Cover + preamble + parties (Issuer / Founders / Investors / round size / minimum cheque / target close)
+2. Share economics & use of proceeds (instrument, valuation, dilution, issue price, share count, use of funds)
+3. Cap table on closing (post-money, fully diluted, total = 100%)
+4. Governance & shareholder protections (board, info rights, pre-emption, drag/tag, SEIS/EIS posture)
+5. Closing terms (governing law, IP assignment, conditions to closing, long-form docs timeline)
+6. Signatures (Company + Investor blocks)
+7. **Appendix A — High Net Worth Individual Investor Statement** (FSMA 2000 / FCA COBS 4 Annex 2 verbatim)
+8. **Appendix B — Self-Certified Sophisticated Investor Statement** (FSMA 2000 / FCA COBS 4 Annex 4 verbatim)
+9. Backpage (contact)
+
+**Doctrine for the term sheet:**
+
+- **Entity name is the IN01-filed name**, not a marketing entity. Issuer must read as the actual filed company (e.g. the name on the IN01 statement of capital, not the trading or marketing brand).
+- **Cap table matches the IN01 + Round 1 dilution model exactly.** Pre-investment shares + new investor Round Shares = post-investment total. No fractional shares; no rounding to the % column that breaks share totals.
+- **Founder list reads exactly as the cap table.** No phantom founders. Strategic shareholders (e.g. an investing LLP) ride in the cap table by entity name, not via their individuals.
+- **Sole director on IN01 is the director-designate.** A term sheet that names two directors when IN01 names one is a defect.
+- **SEIS/EIS posture is post-funding obligation, not closing contingency.** Never assured prior to distribution; no representation as to qualifying status. Forms no part of consideration.
+- **Confidentiality and exclusivity provisions are the only binding clauses.** Everything else non-binding until long-form docs.
+- **FCA self-certification statements are appendices to the term sheet, not separate documents.** The High Net Worth Individual Investor Statement (FSMA 2000 (Financial Promotion) Order 2005 / FCA COBS 4 Annex 2) and the Self-Certified Sophisticated Investor Statement (COBS 4 Annex 4) ship as Appendix A and Appendix B. **Use verbatim wording**; never paraphrase regulatory text. Current source: UKBAA's templated PDFs (Feb 2024 versions reflect the post-31 Jan 2024 reverted thresholds: HNW £170k income or £430k net assets; sophisticated director threshold £1.6m turnover). When the regulatory wording changes, the templates change first at UKBAA / FCA COBS Annexes; re-fetch before issue.
+
+References: `references/termsheet-template.md`.
+
+## Stage 4 — Deck blueprint synthesis
+
+Generate `01-narrative/deck-blueprint.md` as a **13-slide brief** the deck designer can execute. Each slide has Purpose, Headline (one-line message), Body (bullets + figures pulled from the business plan), Visual suggestion, and Source reference. The 13 slides:
+
+1. Cover (raise headline + founders)
+2. Vision (Vision / Mission / Doctrine)
+3. The Market (TAM + tailwind + wedge + **CAGR with citation**)
+4. Business Model (subscription + service layer + unit economics)
+5. Go-to-Market (Wedge → Beachhead → Blitzscale + named partners)
+6. The Product (substrate + wedge + network + asymmetric capture + services layer)
+7. Competition (three failing categories + most-direct counter)
+8. The Team (founders + strategic shareholders)
+9. Economics (P&L summary + cash flow + break-even + **LTV:CAC and burn multiple**)
+10. The Raise (size, valuation, dilution, instrument, use of funds, IP transfer, follow-on round)
+11. Exit (target exit value + 5–8-year horizon + routes — strategic / PE / IPO — + 3–5 named acquirers with one-line rationale)
+12. Milestones (M0 to M-final timeline)
+13. Other Information (compliance + platform dependency + contact)
+
+Production notes attached: format (16:9 landscape, PDF + source deck), design tokens reference (the operator's project design-tokens spec, path supplied at Stage 5 invocation), chart-source consistency with the financial model, speaker-note attribution.
+
+References: `references/deck-blueprint-template.md`.
+
+## Stage 4b — Prospectus synthesis
+
+Generate `html/prospectus/index.html` as an **A4 portrait, 7-page persuasive prospectus** alongside the formal term sheet from Stage 3b. The prospectus is the narrative companion to the term sheet: cover + executive summary + architecture + go-to-market + platform asset + operating plan + use of funds / founders / team.
+
+The templates live under `templates/prospectus/`:
+
+```
+templates/prospectus/
+  index.html          # 7-page prospectus with {{ placeholders }}
+  term_sheet.html     # 8-page term sheet with {{ placeholders }} (shared with Stage 3b)
+  render-pdf.mjs      # image-only PDF renderer (shared with Stage 5)
+```
+
+**Intake — ask these before substituting any placeholder.** Group by section; wait for answers before drafting.
+
+*Company identity:*
+
+1. Legal name of the company (use "to be incorporated" if pre-IN01).
+2. Trading / brand name (topbar + body copy + signatures).
+3. Tagline / one-paragraph deck — what does the company do, who is it for, why is it different. ≤ 80 words.
+4. Brand icon — 22×22 px PNG/SVG mark. Save as `<brand>-icon.png` alongside the HTML.
+5. Cover hero image — landscape photograph, ≥ 2800 px on the long edge. Save as `hero.png` alongside the HTML.
+6. Document month / year (topbar meta, e.g. "May 2026").
+7. Document IDs — short refs for footers: prospectus (e.g. "ACME / FIN-001 / V1.0") and term sheet (e.g. "ACME / TS-001 / V1.0").
+
+*The round:*
+
+8. Round size, gross, £.
+9. Post-money valuation cap, £.
+10. Issue price per share, £. Confirm split between nominal value (typically £0.01) and share premium.
+11. Number of new Round Shares (= round size ÷ issue price).
+12. Founder shares retained at close — total and percentage (typically 80%).
+13. Total shares post-closing.
+14. Minimum cheque per investor, £.
+15. Minimum aggregate raise, £, below which the round does not proceed.
+16. Round close date and funds expected date (DD Month YYYY).
+17. Use-of-proceeds split — typically four buckets (product, GTM, infrastructure, working capital + legal) totalling 100%.
+
+*Founders and team:*
+
+18. Founders — for each: full name, role, share count, percentage. Mark full-time vs advisors.
+19. Director-designates at closing (at least one executive director).
+20. NED nomination right — confirm threshold (e.g. any investor ≥ 10% may nominate one NED, subject to founder approval not unreasonably withheld).
+21. Observer threshold (typically ≥ 5%).
+22. First product hires planned in Y1 — trigger conditions for each.
+
+*The thesis (narrative):*
+
+23. The market — size, growth rate, target customer cohort. Cite sources (ONS, industry reports, named studies).
+24. The opportunity / problem — what is currently broken or unmet.
+25. Value proposition — pricing, what the customer gets, why it beats alternatives.
+26. Architecture and differentiation — the structural moat. What does the company do that nobody else does? "AI" is not a moat; the moat is in *how* AI is applied.
+27. Wedge → beachhead → blitzscale — staged GTM. Wedge = narrow entry product. Beachhead = one decisively winnable segment funded by Round 1. Blitzscale = market expansion funded by Round 2.
+28. Distribution funnel — specific channels producing customers in months 1–6. Concrete: integrations, channels, named partners.
+29. Pilots / paying customers at close — name them, who runs them, what they're paying.
+
+*Financial model:*
+
+30. Opening cash post-close — founder subscription + this round proceeds, less any capex at M0.
+31. Capex at M0 — one-off hardware / set-up costs.
+32. Monthly burn by phase — M1–M2 (founders only), M3–M9 (with first hire), M10–M12 (with second hire). Per-line breakdown.
+33. MRR ramp — month-by-month MRR for M1–M12. Distinguish customer segments.
+34. Customer count and device count by month if hardware is part of the offering.
+35. Cash trajectory — month-by-month closing cash balance derived from (33) − burn. Identify cash low point and break-even month.
+36. Round 2 plan — size range, timing (which month), gating milestones, what it funds.
+
+*Operational / legal:*
+
+37. Founder IP assignment — who is assigning IP to the company at closing? Outright? Any licence-back?
+38. Conditions to closing — incorporation, IP assignment, DD, shareholders' agreement, cleared funds.
+39. Governing law — confirm "English law, Exclusive jurisdiction of England & Wales courts" or substitute.
+40. Long-form documentation timeline — within how many days of mutual signature will the subscription & shareholders' agreement and articles be circulated?
+
+If the operator cannot answer (28), (33–35), or (36), flag explicitly that the prospectus will be incomplete and ask for sensible base-case assumptions. The financial chart on the cover and the operating-plan section both depend on these numbers.
+
+**How to apply answers.**
+
+1. Copy `templates/prospectus/{index.html, term_sheet.html, render-pdf.mjs}` into `html/prospectus/`.
+2. Save the operator-supplied hero image as `hero.png` and brand icon as `<brand>-icon.png`.
+3. Replace **global atomic placeholders** in both HTML files (the templates already carry these): `{{ BRAND_NAME }}`, `{{ COMPANY_LEGAL_NAME }}`, `{{ BRAND_ICON }}`, `{{ HERO_IMAGE }}`, `{{ TERM_SHEET_HTML }}`, `{{ TERM_SHEET_PDF }}`, `{{ PROSPECTUS_PDF }}`, `{{ TERM_SHEET_DOC_ID }}`, `{{ PROSPECTUS_DOC_ID }}`, `{{ DOC_MONTH_YEAR }}`, `{{ ROUND_SIZE }}`, `{{ POST_MONEY_CAP }}`, `{{ ISSUE_PRICE }}`, `{{ NOMINAL_VALUE }}`, `{{ SHARE_PREMIUM }}`, `{{ ROUND_SHARES }}`, `{{ FOUNDER_SHARES }}`, `{{ TOTAL_SHARES_POST }}`, `{{ MIN_CHEQUE }}`, `{{ MIN_AGGREGATE }}`, `{{ ROUND_CLOSE_DATE }}`, `{{ FUNDS_EXPECTED_DATE }}`, `{{ NEW_INVESTOR_PCT }}`, `{{ FOUNDERS_PCT }}`.
+4. Replace **narrative blocks** in `index.html` paragraph-by-paragraph using the answers from sections (23)–(29) and (32)–(36). The placeholder narrative blocks are structural references, not content to keep.
+5. Replace **cap table rows** in `term_sheet.html` (`<tr>` inside `<div class="captable">`) with the operator's actual founder names, roles, share counts, percentages.
+6. Replace **signature block** in `term_sheet.html` with the director-designates.
+7. Replace **financial chart data** in `index.html`. The cover chart is hand-coded SVG — find the `<polyline points="...">` for the cash line, the `<rect>` MRR bars, and the burn step-lines, and recompute coordinates from the new financial model (each MRR bar's `y` = `290 - (mrr_k / 60) * 120`, height = `120 - y_offset`). If recomputation is non-trivial, ask the operator for the 12 monthly cash and MRR values directly.
+8. Update `<title>` and `<meta>` tags to reflect the new company.
+
+**Doctrine for the prospectus.**
+
+- **UK seed round only.** The templates encode UK FCA self-certification under FSMA 2000 (FPO) articles 48 (HNWI) and 50A (Self-certified Sophisticated Investor). Do not adapt for US, EU, or non-UK regimes without explicit operator instruction — the legal exemption text is jurisdiction-specific.
+- **SEIS/EIS is post-funding, not a closing contingency.** The templates state that SEIS/EIS advance assurance will be pursued *after* the round closes and is not a condition of subscription. Do not promote SEIS/EIS as assured before distribution — it is misleading and breaks the FCA exemption.
+- **Image-only PDF via Stage 5's `render-pdf.mjs`.** Same pipeline; the prospectus shares one renderer with the business plan and deck.
+- **A4 fit is enforced.** Every `<section class="page">` must have `scrollHeight ≤ 1123 px + 2 px slack`. The remedy on overflow is to split the section into two `<section class="page">` blocks — never to shrink fonts or pad less.
+- **Render to the same folder as the HTML.** The renderer reads relative asset paths (icon, hero image).
+
+**Verify before shipping.**
+
+1. Read every PNG in `.tmp-*/`. `pdfinfo` reporting "7 pages, A4" is not verification. Confirm: content fills the page, no whitespace gaps > 25 mm at the bottom, the cover hero looks right, the chart values match the model, founder names are spelled correctly, the cap table totals to 100%.
+2. Hero image resolution floor: 2800 px on the long edge. Below that the cover prints soft.
+3. The two FCA self-cert pages render cleanly — HNWI (FPO art. 48) and Sophisticated (FPO art. 50A). These are legally prescribed text; do not edit their wording without lawyer sign-off.
+
+## Stage 5 — HTML + PDF render pipeline
+
+For each artefact that ships as a print-ready PDF (business plan, prospectus, deck), produce three files under `html/<artefact>/`:
+
+```
+index.html        # A4 portrait single HTML file, premium design tokens, sectioned by <section class="page">
+render-pdf.mjs    # Playwright + img2pdf + qpdf pipeline (image-only PDF, linearized)
+<artefact>.pdf    # rendered output (regenerate after every HTML edit)
+```
+
+**Hard constraints (A4 portrait, image-only PDF):**
+
+- A4 portrait: `@page { size: A4 portrait; margin: 18mm 16mm 20mm; }`
+- One `<section class="page">` per A4 page; renderer aborts if `scrollHeight` exceeds 1123px + 2px slack.
+- Page numbers via CSS counter in `.page::after { content: counter(pagenum); }`. The renderer screenshots each section, so `@page @bottom-center` rules do NOT propagate.
+- Image-only PDF: Playwright captures one PNG per section at deviceScaleFactor 3.125 (300 dpi); `img2pdf --pagesize 210mmx297mm` binds them; `qpdf --linearize --object-streams=disable` linearises.
+- Even page count for booklet binding (multiple of 4 ideal). Add or remove a supporting page if parity breaks.
+- Cover and backpage are full-bleed `--surface-deep` with `@page :first { margin: 0 }` and named `@page backpage-full`.
+- Screen-only chrome (e.g. a Download PDF pill button) uses `@media print { display: none !important; }` per the skill.
+- `execFileSync` for img2pdf and qpdf (never `execSync` with template strings: shell-injection class).
+- Verify: `qpdf --check`, `pdfinfo | grep -E "Pages|Optimized"`, **and read every page visually**. `pdfinfo` reporting "25 pages, A4" is not verification.
+
+**Design tokens.** Reference the project's `design-tokens.md` Tier 1 (raw) + Tier 2 (semantic aliases) + Tier 3 (component) layers. Premium register defaults: `--paper-25` near-white page surface, `--teal-900` ink, `--gold-500` accent, `--paper-100` cool mist (`#ECF1F2`) tinted-band surface. Update token values in one place; CSS `:root` mirrors the spec.
+
+**Logo placement.** Copy the light and dark icon variants from the project's public assets (`logo-light.png` for dark cover/backpage, `logo-dark.png` reserved for light content pages if needed) into the artefact directory. Cover top-left pairs the icon with the wordmark; backpage top-left does the same.
+
+References: `references/html-pdf-pipeline.md` for the canonical `render-pdf.mjs` source and CSS scaffold.
+
+## Doctrine (enforced across all stages)
+
+- **No internal workings in public artefacts.** Comparison-to-baseline, reconciliation-pending, open-Q markers, internal cross-references all stay in working files (`01-narrative/office-hours-design.md`, section READMEs). Public artefacts (business plan, deck, prospectus, PDFs) state the current position as a fact.
+- **Every claim cites a source.** Companies House for the cap table. The model worksheet for every financial figure. A named publication for every market figure. A specific vendor page for every competitor claim.
+- **Design tokens are the visual contract.** Update once in the tokens spec; the business plan CSS, the prospectus CSS, and the deck CSS all consume the same tokens. Never hardcode hex values inside component CSS.
+- **Even page count for booklet binding.** Multiple of 4 ideally; minimum even. Add or remove a supporting page to maintain parity if a section change breaks it.
+- **Verify by reading every rendered page.** `pdfinfo` page count is not verification. The bottom 30% of every page must reach the content margin without significant whitespace.
+
+## Slide-design patterns (collected from the deck build)
+
+- **Cap-table over labelled pie chart.** A labelled donut with seven slices puts the percentages on the chart but adds visual clutter at small slice counts. A clean post-investment cap table (Shareholder / Shares / % with a gold-rule Total row) reads faster on the slide and matches the term-sheet treatment. Default to the table; reserve the pie for slides with three to four large slices.
+- **Circular founder avatars on the team slide.** Square photos at full card width over-allocate vertical real-estate and crowd the bio + advisor band. Fixed-diameter circular avatars (e.g. 42mm × 42mm, `border-radius: 50%`, centred `margin: 0 auto`) leave room for the eyebrow + name + 3-line bio + advisor band on a 16:9 slide.
+- **Headshot crops are face-centred, not full-environment.** Crop each source PNG/JPG to a tight head + shoulders square at 1200×1200 (`magick <src> -crop WxH+X+Y +repage -resize 1200x1200 -strip <crop>.png`), then encode with `cwebp -q 88 -m 6 -mt`. Resize-then-encode never loses face fidelity at the 42mm display size. Track per-founder face position from the source dimensions and centre the crop window on it.
+- **Bio line budget on slide 8 is ~3 lines at 8.5pt.** A 4th line crashes into the advisor band's `border-top`. Trim aggressively: lead with the headline number ("33 years residential estate agency"), then 2-3 firms and 1-2 awards.
+- **Linter strips trailing text in long edits.** When editing a multi-line founder bio (or any cell where you append clauses past a verbatim sentence), the auto-formatter has rewritten the trailing fragment on this project. Always re-grep the file after the edit; if "Sole IN01 director." or "2× Capper" went missing, the edit needs a second pass with a smaller anchor.
+
+## Pitfalls to avoid (collected from the reference project)
+
+- **Renaming a directory after the URL is published.** Keep `<project-site>/.docs/data-room/html/<artefact>/index.html` stable; if the URL changes, downstream investor links break.
+- **Letting Round-2 hires inflate the Year-1 P&L.** Only Round-1 hires belong in the Year-1 income statement. Customer-success and product-engineer hires that trigger in Round 1 are in; senior platform engineer, head of growth, and CS team scale that Round 2 funds are not.
+- **Mixing "Year 1" (12-month) and "Round 1 runway" (9-month).** If the £350k buys 9 months, present a 9-month forecast; do not present a 12-month forecast that implicitly assumes Round 2 closes mid-period without modelling the Round-2 cash injection.
+- **Generic AI-bolted-onto-cloud-CRM competitors.** Treat the competitive set as three failing categories: incumbents (cloud CRM + bolted AI), Anthropic horizontal AI (inaccessible to non-tech-wizard users), and new-build cloud-SaaS AI-OS replacements (migration-required, same lock-in pattern as incumbents). Name each vendor, cite each claim from the vendor's own page, rebut with the substrate-AI inversion.
+- **Forgetting Priestley / equivalent founder-thesis anchor.** Any founder who built the same thesis years before AI made it structurally apparent should reference it: book title, first-publication date, founder's costly-signal action (e.g. "bought 200 copies and distributed them to operators in his network"). This dates the conviction and lifts the deck's signal weight.
+- **Letting the page background go light brown.** Premium register page surface (`--paper-25`) is near-white (`#FBFAF6`), not ivory. The screen-preview body background stays plain white (`#FFFFFF`).
+
+## References
+
+- `references/data-room-structure.md` — section-by-section README templates
+- `references/business-plan-template.md` — 15-section business plan template
+- `references/deck-blueprint-template.md` — 12-slide deck brief template
+- `references/html-pdf-pipeline.md` — render-pdf.mjs canonical source + CSS scaffold
+- `references/compliance-research-checklist.md` — UK + EU regulatory regimes to verify
+- `references/internal-workings-scrub.md` — the public-vs-internal doctrine, patterns to scrub, recurrence list
+- `references/design-tokens-application.md` — how to consume the operator's project design-tokens spec from an artefact's `:root` block
+- `references/termsheet-template.md` — A4-portrait term-sheet structure, cap-table reconciliation, common defects
+- `templates/prospectus/` — A4-portrait prospectus + term-sheet HTML templates and the shared `render-pdf.mjs` (consumed by Stages 3b, 4b, 5)
+
+### Analytical frameworks (load on-demand)
+
+Method-only references for the market, financial, and strategy sections. Load the one the live question needs; do not run them for completeness. Each routes every quantitative figure through the `research-assistant` specialist (with `deep-research` / `data-research`) and cites it — no fabricated numbers. Apply `internal-workings-scrub.md` before any output reaches a public artefact.
+
+| Framework | File | Feeds |
+|---|---|---|
+| Pricing strategy | `references/pricing-strategy.md` | `04-financials/`, business-plan §6, prospectus |
+| Monetization strategy | `references/monetization-strategy.md` | `04-financials/`, business-plan §6, prospectus |
+| Competitor analysis | `references/competitor-analysis.md` | `08-market/`, business-plan §8 |
+| Competitive battlecard | `references/competitive-battlecard.md` | `08-market/`, business-plan §7 GTM |
+| Strategy frameworks (SWOT / PESTLE / Porter's / Ansoff) | `references/strategy-frameworks.md` | `08-market/`, business-plan §4/§7/§8, prospectus |
+
+Market sizing (TAM/SAM/SOM) and the pre-fundraise strategy red-team live in the sibling `zero-to-prototype/references/` (`market-sizing.md`, `strategy-red-team.md`) and feed `08-market/` + business-plan §4.
+
+## Reference example
+
+When this skill has previously rendered a data room for the current operator, that directory is the canonical example — read it before drafting. When no prior data room exists for the operator, the fallback reference is `references/data-room-structure.md` (section-by-section purpose statements) plus the per-stage references listed above.

package/payload/premium-plugins/venture-studio/skills/investor-data-room/references/business-plan-template.md

@@ -0,0 +1,64 @@
+# Business plan template — 16-section structure
+
+The canonical structure used by `01-narrative/business-plan.md`. Every public-facing artefact (deck, prospectus, term sheet) derives from this file. Apply the internal-workings scrub (see `internal-workings-scrub.md`) on every revision.
+
+## Section-by-section skeleton
+
+### 1. Executive Summary
+One paragraph each: what the company is (legal entity + Companies House identifier + substrate description), what is being raised (size, pre-money, post-money, runway, Round 2 follow-on), demand evidence (numbers with dates), market size + structural tailwind, competitive set in three failing categories + the architectural inversion, bottom-line P&L summary.
+
+### 2. Vision and Mission
+Three blocks. **Vision.** Where the company ends up at full execution. **Mission.** What the company gives to its customer day one. **Doctrine.** The structural belief that everything else derives from.
+
+### 3. The Company
+Legal entity table (legal name, Companies House number, incorporation date, registered office, SIC, PSC status, sole proposed director on IN01) + pre-investment cap table (sourced from the IN01 verbatim) + post-investment dilution pointer.
+
+### 4. The Market
+Subsections: Size of the addressable market (TAM with citation per figure, bottoms-up construction; geography-tier floor — **late pre-seed TAM is expected to clear $1B US / $500M UK / $500M EU**, flag if the named geography sits below its floor); Market growth rate (single CAGR figure with citation + a one-line 3-year market-expansion thesis stating why the market gets bigger and why this company captures more share as it does); Market-share trajectory (path to 3-8% of SAM within 5 years, reconciled line-by-line against the ARR target in Section 12); Structural tailwind (margin squeeze, regulatory tailwind, or equivalent); Self-employed wedge or analogous high-growth segment; Domestic adjacency (next-vertical expansion); International upside (channel partner enabling it); Pricing and TAM (per-customer ACV × addressable count = TAM number).
+
+### 5. The Product
+Subsections: What it is (substrate description, CRM-agnostic / data-sovereign positioning); The wedge product (narrowest unit of value, public-by-default if applicable); The upsell path (wedge → high-value output → full substrate); The substrate moat (operator-data + ontological graph + flywheel); Conversation ingestion (or equivalent lock-in mechanic); The network (federation thesis); Asymmetric-capture position (the LLM-binary wrap argument, with commercial-practice insurance); Human-services layer (training, support, white-glove, anti-fragile-against-frontier-AI).
+
+### 6. The Business Model
+Revenue streams (subscription tiers + future network-effect revenue). Unit economics — each metric is a named line: **ACV**, **CAC**, **LTV**, **LTV:CAC ratio** (target ≥ 3:1 by Month 12), **gross margin** (target ≥ 70% steady-state), **burn multiple** (net burn ÷ net new ARR; target ≤ 2× during wedge, ≤ 1× post-PMF), **churn** (logo + revenue). Compute-mark-up doctrine ("never mark up the LLM compute; customer pays the LLM provider directly").
+
+### 7. Go-to-Market
+Wedge → Beachhead → Blitzscale framework (with months and funding source per phase), demand evidence (40+ EOIs, paid intents, founder dogfood, partnership status — every number cited), GTM partners (one row per named partner with role and source), horse-before-cart logic (public-first wedge that earns trust for the private high-value upsell), geography.
+
+### 8. Competition
+Three failing categories: (a) cloud-CRM-bolted-AI incumbents, (b) horizontal-AI providers inaccessible to ICP, (c) new-build cloud-SaaS replacements requiring full migration. RANL differentiation paragraph. Most-direct-competitor counter (CRM-agnostic vs migration-required is the headline rebuttal).
+
+### 9. Operations
+Technology stack (foundation model, graph store, hardware tiered by workload, network, hosting), data sovereignty paragraph, IP ownership chain (deed-of-assignment commitment), compliance posture operational summary (pointer to Section 10 for full treatment).
+
+### 10. Compliance and regulatory positioning
+Three regulatory regimes: UK consumer protection (DMCC Act for estate agency / equivalent for the vertical), UK data and AI regime (UK GDPR + ICO ADM Articles 22A–22D), EU AI Act (relevant only to international expansion). Adjacent UK obligations (Estate Agents Act / equivalent + AML + sector membership bodies). Closing: "Why compliance is the moat, not the tax".
+
+### 11. Team
+Founders (foregrounded) with role + concise career summary + key credentials + pre-investment shareholding. Strategic shareholders from incorporation (advisors, family, JV partners on the cap table from day one). Hiring plan table with month-triggered roles + salary + trigger condition.
+
+### 12. Financial Plan
+12.1 Opening balance sheet (M0 after capex). 12.2 Year-1 Income Statement (revenue + COGS + every opex line + depreciation + EBIT + corp tax + net loss). 12.3 Year-1 cash flow summary table (period-by-period opening, net flow, closing cash). 12.4 Closing balance sheet (M12). 12.5 Two-round funding strategy (timing, size, pre-money, purpose, gating evidence per round) — **valuation per round cites `03-cap-table/comparables.md`** (recent comparable rounds: company / pre-money / post-money / multiple / source). 12.6 Contingency levers. 12.7 Use of funds (this round) by bucket.
+
+### 13. Risks and Mitigations
+Two-column table: risk + mitigation. Cover Loop / Anthropic / competitor-ships-earlier / wedge-fails-to-convert / founder-transition / counsel-delays / Round-2-market / investor-structure.
+
+### 14. Milestones (the N-month thesis)
+Month-by-month table: M0 close + closing conditions; M1 first paying customer; M3 product-engineer hire + ~10 customers; M6 wedge phase complete; M7 Round 2 raise opens; M9 Round 2 closes; M12 break-even.
+
+### 14b. Exit thesis
+Target exit value (£m), horizon (5–8 years), exit routes considered (strategic acquisition, PE rollup, IPO) with a one-line view on which route is most likely and why. **3–5 named acquirers**, each with a one-line rationale (why they would buy: distribution gap, technology gap, defensive play, vertical entry). Sourced from `08-market/sources.md` and `03-cap-table/comparables.md` (recent exit comparables: company / acquirer / multiple / year / source). Body references "Appendix B" by title.
+
+### 15. The bottom line
+Summary table of headline numbers (raise size, pre-money, post-money, dilution, runway, Y1 revenue/loss/cash-burn/closing-cash/ARR, break-even month, Round 2 timing/size). Closing paragraph.
+
+### Appendices
+Cited documents listed by title only. Each available to a serious investor on request; internal-only working documents are not listed.
+
+## Writing rules
+
+- **Numbers cite their source.** Every figure in the body links to an Appendix entry that is itself a publicly-shareable document.
+- **No internal commentary.** See `internal-workings-scrub.md`.
+- **No em-dashes between alphabetic words** in user-facing prose. Use colons, parentheses, semicolons, or rewrites.
+- **Even-page-count parity** if the document is printed; add a supporting page if a section change breaks parity.
+- **One pre-investment cap table source: the IN01.** Reconcile the nominal-value discrepancy (e.g. £0.001 vs £0.01) before writing the balance sheet.

package/payload/premium-plugins/venture-studio/skills/investor-data-room/references/competitive-battlecard.md

@@ -0,0 +1,37 @@
+# Competitive Battlecard
+
+The method for a sales-ready one-pager per key competitor — the fast-reference an operator (or a founder in a sales call) uses to position against a named alternative. Built from the analytical teardown in `competitor-analysis.md`. Output lands in `08-market/` and supports the go-to-market narrative (business-plan §7).
+
+## What this produces
+One card per priority competitor: who they are, where they win, where they lose to you, the objection-handlers, and the trap questions that expose their weakness — all grounded in the sourced facts from `competitor-analysis.md`. Apply `internal-workings-scrub.md` to anything that reaches a public-facing document; battlecards are usually internal, but the facts must still be sourced.
+
+## Citation discipline (binding)
+A battlecard is only useful if its claims are true. Every factual line — their pricing, their limitation, their funding — traces to the sourced profile in `competitor-analysis.md`, which routes through the `research-assistant` specialist with `deep-research`. Do not put a claim on a card you cannot source; mark gaps `source pending — route to research-assistant`. A card that arms a founder with a wrong fact loses the deal.
+
+## Process
+
+### 1. Pick the competitors worth a card
+Only the alternatives you actually lose deals to or get compared against. Three to five cards beat fifteen. The status quo ("they'll just keep doing it manually") deserves a card.
+
+### 2. Build the card on a fixed template
+Each card, same shape:
+- **Who they are** — one line, sourced.
+- **Where they win** — the honest reasons a buyer picks them (credibility matters; pretending they have no strengths loses trust).
+- **Where they lose to us** — the structural weakness from the teardown, stated as a buyer outcome.
+- **Objection handlers** — the two or three things their salesperson says about you, and the true one-line response to each.
+- **Trap questions** — questions the buyer can ask the competitor that expose the weakness (e.g. "can I take my data and leave?").
+
+### 3. Keep it to outcomes, not features
+A battlecard sells the buyer's outcome, not your feature list. "They lock your data in their cloud; you can't leave without losing it" beats "we support data export".
+
+### 4. Keep it current
+Competitor facts go stale. Each card carries the source so it can be re-checked; when a competitor's pricing or capability changes, re-run the `research-assistant` lookup and update the card.
+
+## Output
+`08-market/` working notes, one card per priority competitor, each tracing its facts to `competitor-analysis.md`. Supports business-plan §7 GTM and live sales positioning.
+
+## Common failures
+- Pretending the competitor has no strengths — buyers see through it.
+- Feature-versus-feature instead of outcome-versus-outcome.
+- Stale facts with no source to re-check against.
+- A card for every competitor instead of the few that matter.

package/payload/premium-plugins/venture-studio/skills/investor-data-room/references/competitor-analysis.md

@@ -0,0 +1,32 @@
+# Competitor Analysis
+
+The method for a structured competitor teardown that feeds the business plan's competition section. Output lands in `08-market/` and feeds business-plan §8 "Competition". This is the analytical backbone; the sales-facing one-pager built from it is `competitive-battlecard.md`.
+
+## What this produces
+A competitor set grouped by **how they fail the buyer**, each with a sourced profile, and a differentiation paragraph stating the architectural inversion that the incumbents structurally cannot copy. Apply `internal-workings-scrub.md` before anything reaches a public document.
+
+## Citation discipline (binding)
+Every factual claim about a competitor — pricing, funding, customer count, feature presence, market position — is a **sourced claim**. Route competitor facts through the `research-assistant` specialist with `deep-research`; cite each. Do not assert a competitor "can't do X" or "charges Y" from memory. Unsourced claims are written `source pending — route to research-assistant`. A teardown built on remembered facts is a liability in diligence — competitors read these.
+
+## Process
+
+### 1. Define the competitive set by failure mode
+Don't list competitors flat — group them by the **category of how they fail the buyer** (the data-room's standing structure uses three failing categories, e.g. incumbents with bolted-on AI, horizontal providers inaccessible to the ICP, full-migration replacements). Each group shares a structural weakness you exploit. Include the status quo ("doing nothing") as a competitor — it is usually the real one.
+
+### 2. Profile each competitor on a fixed axis set
+For comparability, profile every competitor on the same axes: what they do, who they serve, sourced pricing, sourced funding/scale, the specific way they fail your buyer, and switching cost for their customers. Same axes for all → a real comparison, not a brochure.
+
+### 3. State the differentiation as an inversion
+The differentiation paragraph names the **structural** reason you win — the architectural choice (data sovereignty, CRM-agnosticism, public-first wedge, etc.) that the incumbent cannot match without breaking their own model. "We're better/faster/cheaper" is not differentiation; "they cannot do this without abandoning their core business" is.
+
+### 4. Counter the most-direct competitor explicitly
+Name the single closest alternative and write the head-to-head rebuttal — the one sentence an investor will test you on. Source the claim it rests on.
+
+## Output
+`08-market/` working notes → business-plan §8. Format: the failure-mode groupings, a fixed-axis profile per competitor with citations, the inversion differentiation paragraph, and the direct-competitor counter.
+
+## Common failures
+- A flat feature-grid with no failure-mode structure.
+- Competitor facts from memory — wrong and dangerous.
+- "Differentiation" that is just adjectives, not a structural inversion.
+- Omitting the status quo as a competitor.

package/payload/premium-plugins/venture-studio/skills/investor-data-room/references/compliance-research-checklist.md

@@ -0,0 +1,53 @@
+# Compliance research checklist
+
+Required before drafting business plan Section 10 (Compliance and regulatory positioning) for a UK-domiciled raise. Every claim must cite a primary source (ICO guidance, gov.uk, regulator strategy document) verified by WebSearch.
+
+## UK consumer-protection regime
+
+**For estate-agency / property:** Digital Markets, Competition and Consumers Act 2024 (DMCC Act). In force from 6 April 2025. Supersedes the Consumer Protection from Unfair Trading Regulations 2008 for estate-agency consumer protection. Under the DMCC Act, **omitting material information is automatically an unfair commercial practice regardless of whether the omission affected the consumer's decision**. CMA holds direct enforcement powers including fines without court process. Property brochures must include Part B material information.
+
+**For other verticals:** Identify the equivalent UK consumer-protection regime (financial services → FCA Consumer Duty; healthcare → CQC; legal services → SRA; etc.). Same pattern: cite the binding statute, name the regulator, identify the enforcement mechanism.
+
+## UK data-protection and AI regime
+
+- **UK GDPR** sits over UK data processing. Articles 22A to 22D (added by the Data Use and Access Act) govern solely automated decision-making.
+- **No UK AI-specific statute in the 2026 parliamentary session.** Labour's 2024 King's Speech committed to "appropriate legislation on the most powerful AI models" but the Bill has not been introduced. A dedicated 2026 AI Bill is not expected.
+- **ICO strategy.** June 2025 strategy "Preventing Harm, Promoting Trust" frames the regulator's approach. ICO consulting on updated Automated Decision-Making and Profiling guidance during 2025–2026, with a statutory code of practice on AI and ADM in the pipeline. First substantive enforcement guidance expected later in 2026.
+- **UK is not adopting the EU AI Act.** Sector-led approach via existing regulators. Starmer's stated position: *"instead of over-regulating these new technologies, we're seizing the opportunities they offer"*.
+
+## EU AI Act (relevant only to international expansion)
+
+- Entered into force 1 August 2024.
+- Fully applicable 2 August 2026.
+- Classifies AI systems used in credit and insurance (which overlap with mortgage and property-financing decisions) as **high-risk**. Most marketing and operational AI applications sit at lower risk tiers.
+- High-risk obligations: risk-management systems, data-quality controls, human oversight, transparency to affected individuals, conformity assessment, post-market monitoring.
+- For UK-domiciled companies, the EU AI Act binds only when serving EU customers. Document the international expansion path and confirm whether any planned product crosses into high-risk territory.
+
+## Adjacent UK obligations to verify per vertical
+
+- **Estate Agents Act 1979** for property: distinguish software vendor (out of scope) from estate agent (in scope).
+- **Money Laundering Regulations 2017 + Economic Crime and Corporate Transparency Act 2023** for AML obligations.
+- **PECR (Privacy and Electronic Communications Regulations 2003)** for electronic marketing, cookies, and tracking. Sits alongside UK GDPR. Enforced by ICO with fines up to £500,000.
+- **Sector membership bodies** (NAEA / ARLA Propertymark for property; equivalents per vertical).
+- **Client Money Protection schemes** where the business handles client money.
+
+## WebSearch protocol
+
+For each regime, search for:
+
+1. The primary statute / regulation page on gov.uk or the regulator's site.
+2. A recent (within last 12 months) legal-firm briefing summarising current state.
+3. The regulator's strategy document.
+
+Examples that worked in the reference project:
+
+- DMCC Act + Propertymark guidance withdrawal: `https://www.propertymark.co.uk/resource/material-information-guidance-withdrawn-as-the-digital-markets-competition-and-consumers-act-takes-over.html`
+- ICO AI strategy: `https://ico.org.uk/for-organisations/uk-gdpr-guidance-and-resources/artificial-intelligence/guidance-on-ai-and-data-protection/`
+- UK AI Bill status: `https://iapp.org/news/a/king-s-speech-signals-diffuse-uk-digital-policy-agenda-but-no-ai-bill`
+- EU AI Act compliance for UK businesses: `https://www.rmoklegal.com/guides/eu-ai-act-compliance-uk`
+
+Cite the actual source URLs in the business plan's reply / Sources block (per the WebSearch tool's mandatory citation rule).
+
+## The closing argument
+
+The business plan's Section 10 ends with "Why compliance is the moat, not the tax": cloud-CRM-bolted-AI competitors inherit the vendor's data-protection posture; the product owns the agent's data sovereignty by design, so compliance is a competitive moat (especially into EU markets where the EU AI Act amplifies the cost differential).