Chronograph LP

View saved version →

---
name: chronograph-cashflow-forecast
description: Forecast private capital cashflows for existing portfolios using Chronograph MCP data and a Takahashi-Alexander style model. Use when Codex needs to analyze or forecast LP-level contributions, distributions, NAV, unfunded exposure, net cashflows, or Excel-style cashflow forecast outputs from existing Chronograph funds, commitments, groups, or portfolios.
---

# Chronograph Cashflow Forecast

**Requirements:** The Existing Portfolio Forecast mode requires a connected Chronograph MCP server — these workflows are designed for permissioned Chronograph users to connect to their private investment data. The Future Commitment Pacing Overlay mode runs from user-provided assumptions and works without a connection.

## Overview

Use the Chronograph MCP server as the source of truth for private capital commitment data, then forecast fund and portfolio cashflows with a Takahashi-Alexander style model. Prefer concise analysis in chat unless the user asks for a workbook, export, or model artifact.

Do not hard-code tool names — read the Chronograph MCP server's live tool descriptions to pick the appropriate tool for each data need.

## Workflow

1. Clarify scope only when needed: portfolio, group, fund IDs, commitment IDs, currency, as-of date, forecast horizon, and units. V1 is existing-portfolio-only; do not model future commitments or pacing schedules.
2. Resolve entities. The Chronograph MCP server exposes a tool for resolving fund, group, GP, or company names to IDs. Use it to turn user-provided names into IDs before pulling any metrics.
3. Pull fund metadata. The Chronograph MCP server exposes a generic query tool for retrieving core entity attributes — use it to get fund name, fund type, vintage year, reporting currency, geographic focus, general partner, and final reporting date where useful.
4. Pull net LP commitment values: NAV, called, distributed, unfunded, commitment amount, net IRR, net MOIC, DPI, and RVPI. The Chronograph MCP server exposes a dedicated tool for net LP-level performance that covers all of these. Treat the values as net and surface the as-of date, currency, and that the values are net (not gross) in user-facing results.
5. Map fund types to forecast assumptions. Use the default assumptions in `references/model-methodology.md` unless the user provides custom assumptions.
6. Build yearly or quarterly forecast periods from the as-of date through the requested horizon. Default to annual periods for executive analysis and quarterly periods for Excel-style output.
7. Forecast contributions, distributions, NAV, unfunded, and net cashflow by fund, then aggregate to portfolio, group, vintage, fund type, or GP as requested.
8. Present assumptions, source context, and checks. Always state currency, units, as-of date, forecast horizon, and that future commitments are excluded in V1.
9. If the user requests Excel, create a workbook with Inputs, Fund Forecast, Portfolio Summary, and Checks.

## Planning Mode

Use one of these modes, or combine them when the user asks for a liquidity plan:

- **Existing Portfolio Forecast.** Use Chronograph actuals for current commitments, NAV, unfunded, called, distributed, and net performance, then forecast runoff for existing commitments only.
- **Future Commitment Pacing Overlay.** Use user-provided planned commitments and strategy assumptions to estimate future calls, distributions, NAV, unfunded, exposure, and net cashflow. These values are planning assumptions, not Chronograph actuals.
- **Combined Liquidity Plan.** Add the existing portfolio forecast and pacing overlay to show total expected calls, distributions, net cashflow, NAV, unfunded, and exposure.

## Chronograph MCP Usage

The skill needs three categories of data from the Chronograph MCP server: entity identifiers (fund and group IDs), fund metadata (name, type, vintage, reporting currency, geographic focus, GP, final reporting date), and net LP-level cashflow values (NAV, called, distributed, unfunded, commitment amount, net IRR, net MOIC, DPI, RVPI). Inspect the server's live tool descriptions and pick the appropriate tool for each category. Do not hard-code tool names.

**Net vs gross — confirm with the user when ambiguous.** This skill forecasts *net* LP cashflows. The Chronograph MCP server exposes separate tools for net LP-level performance and gross fund-level returns. User requests like "show me MOIC for these funds" are ambiguous — confirm net or gross before picking a tool. Gross returns (gross IRR, gross MOIC, cost, realized, unrealized) are not used in this forecast and live in a different tool.

**Currency.** Do not default currency to USD. For specific funds, resolve reporting currency from fund metadata before pulling cashflow values. If reporting currencies conflict across a multi-fund portfolio, ask the user.

**Custom investment-level metrics.** If a tool requires a discovery / help / schema-introspection call first (common pattern on the Chronograph MCP server), make that call to see which metric types are available, then query each line item.

See `references/chronograph-mcp-map.md` for data-need to tool mapping guidance.

## Error Handling

- **MCP not connected.** Depends on mode. For an Existing Portfolio Forecast, stop — the connection is required, because forecasting runoff of a real portfolio needs its actual NAV, unfunded, called, and distributed baseline; do not invent those values or forecast a real portfolio from assumptions alone. For a Future Commitment Pacing Overlay (or a from-scratch pacing request with no existing portfolio), proceed from user-provided commitment amounts, vintages, and strategy assumptions, and label every value as a planning assumption rather than a Chronograph actual. If the user wants a Combined Liquidity Plan but Chronograph is unavailable, produce the pacing overlay only and list the existing-portfolio baseline as a missing input.
- **Entity not found.** Ask the user to confirm the fund / group / GP name. Do not silently substitute a near-match.
- **Currency cannot be resolved.** Ask the user which currency to use rather than defaulting to USD.
- **Future commitments / pacing overlay.** Chronograph is still the source of truth for the existing portfolio baseline. Future commitments are user-provided planning assumptions. Do not imply that planned commitments, future calls, future distributions, future NAV, or wind-down schedules are sourced from Chronograph if they are guided by the user in conversation.
- **Net-vs-gross ambiguity.** Confirm with the user before proceeding.
- **Pacing details incomplete.** If the user gives a high-level plan such as "deploy $500mm for a couple years, then wind down," make a conservative explicit schedule and label it as an assumption. Ask when the missing detail materially changes the decision, such as currency, annual versus total commitment budget, or strategy mix.

## Forecast Method

Use `references/model-methodology.md` for the distilled forecast methodology. Load it when doing actual forecast math, explaining assumptions, building an Excel output, or debugging differences versus a reference workbook.

Core outputs:

- Contributions: forecast capital calls.
- Distributions: forecast realizations/income.
- NAV: forecast residual value.
- Unfunded: forecast remaining callable commitment.
- Net cashflow: distributions minus contributions.
- Total exposure: NAV plus unfunded.

## Output Standards

- Include a compact assumption table whenever forecast values are shown.
- For planning-mode outputs, include a baseline section, an assumed commitment schedule, an overlay cashflow table, a combined liquidity view when available, and caveats / missing inputs.
- Separate actuals from forecasts. Label actual periods with `A` and forecast periods with `E` when producing Excel-style tables.
- Show totals and key years first, then detail by fund or fund type.
- Include basic checks: beginning NAV plus contributions less distributions plus growth should reconcile to ending NAV; prior unfunded less contributions should reconcile to ending unfunded; net cashflow should equal distributions less contributions.
- Avoid overstating precision. Forecasts are model estimates driven by assumptions, not Chronograph-reported future values.

## Guardrails

- Treat the output as draft analyst work product for liquidity and pacing planning.
- Do not provide investment, legal, tax, audit, or valuation advice. A forecast or liquidity plan is a scenario estimate, not a recommendation to commit, call, distribute, or sell.
- Do not present a pacing figure or commitment range as advice to act. Frame it as what the model implies under the stated assumptions, and surface the assumptions that drive it.
- Forecasts are scenario estimates under explicit assumptions, not predictions. Disclose the assumption set and that actual outcomes will differ.
- Do not imply that planned commitments, future calls, future distributions, or future NAV are sourced from Chronograph when they are user-provided planning inputs.

## Excel Outputs

When asked for Excel, use the spreadsheet skill and build a clean workbook rather than recreating the original workbook cell-by-cell. A good workbook shape is:

- `Inputs`: scope, as-of date, currency, units, horizon, assumptions by fund type.
- `Fund Forecast`: one row per fund/period with actuals and forecast values.
- `Portfolio Summary`: annual totals, net cashflow, NAV, unfunded, exposure, charts when useful.
- `Checks`: reconciliation checks and source context.

## Reference Files

- `references/model-methodology.md`: forecast formulas and default assumptions based on the Excel model.
- `references/chronograph-mcp-map.md`: Chronograph MCP data-need to tool mapping guidance for this workflow.

Referenced files: 3

View saved version →

---
name: chronograph-gp-meeting-prep
description: >
  Prepare an LP to meet with their fund manager (GP): review the fund's latest
  reporting, surface what changed since last period, and draft the questions
  worth raising. Use when someone is getting ready for a manager call, quarterly
  check-in, annual meeting, LPAC meeting, or a re-up decision — for example "I
  have my quarterly call with this manager next week, help me prep," "what
  changed in this fund this quarter," "what should I ask them about these marks,"
  or "what are the red flags in this reporting package." Draws on Chronograph
  fund and portfolio data when connected — captured reporting such as fund
  performance, schedules of investments, and portfolio company KPI profiles —
  and asks the LP to provide anything else it needs, such as a capital account
  statement, investor letter, or AGM or board deck.
---

# Chronograph GP Meeting Prep

**Requirements:** A connected Chronograph MCP server. These workflows are designed for permissioned Chronograph users to connect to their private investment data.

## Overview

Prepare an LP for a meeting with a GP by reviewing fund reporting, surfacing what changed, identifying diligence questions, and producing a concise meeting brief. The brief exists to sharpen a forward decision — a re-up, a secondary sale, or an LPAC vote — so prioritize what bears on those. Use Chronograph as the source of truth where connected and permissioned; otherwise work from user-provided reporting materials.

Default to a chat brief. Create a document, workbook, or deck only when the user asks for an artifact.

## Workflow

1. Clarify scope only when needed: GP, fund, reporting period, meeting type, as-of date, currency, units, output format, and whether the user wants a quick brief or a formal prep document.
2. Resolve the relevant fund, manager, commitment, reporting period, and LP perspective from Chronograph where available. If Chronograph is unavailable, ask for the GP reporting package or a pasted/uploaded summary.
3. Gather the fund baseline, anchoring first on the figures the GP reported this period and citing the source document where available: gross performance, cost, realized and unrealized value, current valuation, portfolio holdings, prior-period values, and fund attributes. Carry net performance — NAV, calls, distributions, TVPI/DPI/RVPI, and net IRR — as the LP's headline return, clearly labeled gross versus net and dated.
4. Review document-backed reporting content. When Chronograph document semantic search is available, use it as a primary step to find the story behind the numbers: what drove calls and distributions, how the GP describes its marks, and the commentary on portfolio company KPIs.
5. If a document the meeting needs isn't in the connected context, ask the LP to provide it. Do not infer narrative, management explanations, or meeting context from structured data alone.
6. Compare current period to prior period: performance movement, cashflow activity, NAV and unfunded movement, valuation changes, portfolio company adds/exits, concentration changes, and KPI movements.
7. Separate facts from interpretation. Use reporting data for facts, cite document support when available, and label inferred questions as analyst prompts rather than confirmed issues.
8. Produce a prioritized meeting brief with key changes, watch items, suggested GP questions, and follow-up requests.

## Chronograph Data Use

When the Chronograph MCP is available, use it to retrieve the relevant portfolio, fund, commitment, performance, exposure, and report-derived document context where permissioned. Keep public-facing explanations at the workflow level; do not expose internal schemas, private tool names, field mappings, or retrieval recipes.

Source discipline:

- Cite Chronograph, document, page, period, or user-provided source for quantitative claims whenever available.
- State currency, units, as-of date, and whether values are GP-reported, Chronograph-derived, user-provided, or estimated.
- Do not fabricate NAVs, KPIs, performance metrics, holdings, explanations, or manager commentary.
- If sources disagree, present the discrepancy and ask the GP-facing question that follows from it.

## Review Priorities

Use `references/report-review-checklist.md` for a fuller checklist when reviewing a package. Anchor the review on the figures the GP reported this period, citing the source document where available, and present net performance as the LP's headline return clearly labeled gross versus net. Prioritize:

- Performance: lead with the GP-reported figures — gross IRR and MOIC, cost, and realized and unrealized value — and their movement versus prior period, citing sources where available. Carry net performance (TVPI, DPI, RVPI, net IRR) as the LP's bottom-line return, labeled gross versus net, and watch the trend from unrealized value toward realized distributions.
- Cashflows: calls, distributions, recallable amounts, unfunded exposure, and unusual activity. Distinguish realized-cash distributions from those financed by NAV or subscription-line facilities.
- Valuation: write-ups, write-downs, stale marks, methodology changes (including any shift toward a cost basis), the GP's written valuation policy, and valuation concentration.
- Portfolio: new investments, exits, major company KPI changes, top holdings, sector/geography exposure, and single-name concentration.
- Reporting quality: missing explanations, inconsistent values, stale dates, changed definitions, or unexplained metric movement.
- Meeting readiness: three to eight prioritized questions, required follow-up materials, and open items to track.

## Output

For a quick answer, use this shape:

- Executive snapshot
- What changed this period
- Items to ask the GP
- Follow-up requests
- Source notes and assumptions

For a formal memo or document, use `references/meeting-brief-template.md`.

Use `references/question-bank.md` when the user wants more complete question coverage by topic or when the review surfaces a broad set of issues.

## Fallback Behavior

If Chronograph is unavailable, or the meeting needs a document outside the connected context:

- Ask the LP for the relevant GP reporting package or materials — for example the quarterly report, schedule of investments, capital account statement, or prior meeting notes.
- If only partial data is available, produce a partial brief and list the missing inputs needed for a complete meeting prep.
- If the user provides PDFs, workbooks, or pasted text, extract only what is supported by the material and cite the source artifact.

## Guardrails

- Treat the output as draft analyst work product for LP meeting preparation.
- Do not provide investment, legal, tax, audit, or valuation advice.
- Do not imply that a missing GP explanation is misconduct; frame gaps as questions or follow-up requests.
- Do not infer manager intent or portfolio company facts from metric movement alone.
- Avoid overprecision. Round appropriately and disclose incomplete source coverage.

Referenced files: 4

View saved version →

---
name: chronograph-portfolio-company-one-pager
description: >
  GP platform one-pager and investor report generator for private equity portfolio companies.
  Use this skill whenever a user asks to generate a company tearsheet, one-pager, investor
  report, portfolio overview, or company deep-dive — especially when they name a company or
  ask to "build a report", "create a one-pager", or "show me a tearsheet". Also trigger when
  the user asks to include commentary, quarterly updates, investment narratives, or any
  Investment Overview in the report output. This skill handles live data fetching via a
  connected MCP data source OR from an uploaded Excel model, metric formatting, AI-generated
  or model-sourced commentary, and rendering a fully styled HTML one-pager. Also trigger for
  LP quarterly updates, valuation summaries, and portco performance pages — any output that
  combines financials, valuation, and return data for a single portfolio company.
---

# GP Report Builder

Generates a fully styled, self-contained HTML investor report for a named portfolio company.
Supports two data source modes, two report types, and automatic brand detection from uploaded
templates. **Read this skill fully before writing a single line of HTML.**

**Requirements:** A connected Chronograph MCP server. These workflows are designed for permissioned Chronograph users to connect to their private investment data, or an uploaded Excel model for Model mode.

---

## Step 0 — Resolve Branding

Brand tokens are resolved automatically — the user never fills in a config file manually.
Execute the following decision tree before any other step.

### 0A — Check for an Uploaded Brand Template

Look for any of the following in the current conversation:

| File type | What to extract |
|---|---|
| **HTML / CSS file** | Parse all `color`, `background`, `font-family`, and `border` declarations. Identify the dominant background, primary accent, heading font, and body font. Extract any `--variable` tokens if a design system is present. |
| **PDF report or one-pager** | Visually analyse the document. Identify background color(s), the dominant accent/highlight color, heading and body typefaces, logo presence, and footer text. |
| **Image (PNG / JPG / SVG)** | Extract the 3–5 most visually prominent colors using the image content. Identify any visible text to infer font style (serif vs sans-serif, bold vs light). Note logo or wordmark if present. |
| **PowerPoint / PPTX** | Read slide backgrounds, title font, body font, accent colors from shapes and highlights. |
| **CSS / design token file** | Map token names to the brand token schema below directly. |

Once extracted, map findings to the Brand Token Schema in **0C**.

If the user has dropped **multiple** files, treat the most recent one as authoritative for
branding, unless the user specifies otherwise.

### 0B — No Template Provided → Apply Chronograph Defaults

If no brand template is present in the conversation, apply the following defaults silently —
do not ask the user to provide branding.

```
firm_name:             Chronograph
website:               www.chronograph.pe
confidentiality_label: CONFIDENTIAL

colors:
  bg_primary:       #101C1D   ← Near Black 1 (page background)
  bg_secondary:     #1A2627   ← Near Black 2 (card / panel background)
  bg_header:        #1B4147   ← Deep Teal (header strip)
  accent_primary:   #57E5EE   ← Bright Teal (headlines, KPI values, eyebrows)
  accent_secondary: #11A8B2   ← Regular Teal (bars, borders, left-rule accents)
  accent_negative:  #F95532   ← Accent Red (EBITDA bars, negative deltas)
  accent_positive:  #4ecb8a   ← Green (positive deltas)
  text_primary:     #FFFFFF   ← White (body text)
  text_muted:       #D9E8E8   ← Light teal tint (footnotes, commentary)
  table_header_bg:  #1B4147   ← Deep Teal (table header row)

fonts:
  heading:         Ubuntu
  heading_weight:  700
  body:            Open Sans
  body_weight:     300
  google_fonts_url: https://fonts.googleapis.com/css2?family=Ubuntu:wght@700&family=Open+Sans:wght@300;300i&display=swap

logo:
  url:      (none — render firm name as text)
  position: top-left

theme: dark
```

### 0C — Brand Token Schema

Whether tokens are extracted from a template (0A) or defaults are applied (0B), resolve all
of the following before proceeding. Every downstream panel references these names.

| Token | Role | Fallback if undetectable |
|---|---|---|
| `firm_name` | Firm name in header and footer | Infer from logo text or filename; else `"Your Firm"` |
| `website` | Footer URL | `""` (omit from footer) |
| `confidentiality_label` | Footer suffix | `"CONFIDENTIAL"` |
| `bg_primary` | Page / root background | Chronograph default |
| `bg_secondary` | Card / panel background | Darken `bg_primary` by 5% |
| `bg_header` | Header strip background | Darken `bg_primary` by 15% |
| `accent_primary` | Headlines, KPI values, eyebrow labels | Dominant bright color from template |
| `accent_secondary` | Bars, borders, left-rule accents | Mute `accent_primary` by 30% |
| `accent_negative` | Negative values, downward deltas | `#F95532` |
| `accent_positive` | Positive deltas | `#4ecb8a` |
| `text_primary` | Main body text | `#FFFFFF` on dark; `#1A1A1A` on light |
| `text_muted` | Footnotes, commentary | Tint `text_primary` toward `bg_primary` by 20% |
| `table_header_bg` | Table header row | `bg_header` |
| `font_heading` | Heading / KPI / eyebrow font | Detected from template; else `Ubuntu` |
| `font_heading_weight` | Heading weight | `700` |
| `font_body` | Body / table / footnote font | Detected from template; else `Open Sans` |
| `font_body_weight` | Body weight | `300` |
| `google_fonts_url` | Font load URL | Build from detected font names |
| `logo_url` | Logo image src | `""` — fall back to firm name as text |
| `logo_position` | Header logo placement | `top-left` |
| `theme` | `dark` or `light` | `dark` if `bg_primary` luminance < 0.2; else `light` |

### 0D — Theme Handling

**Dark theme** (luminance of `bg_primary` < 0.2):
Use the token values as resolved. Default contrast pairings apply (white text on dark
backgrounds). Minimum contrast ratio: 4.5:1 for body text, 3:1 for large text (≥ 18px bold).

**Light theme** (luminance of `bg_primary` ≥ 0.2):
- Swap `text_primary` to a near-black (e.g. `#1A1A1A`) if not already dark
- Ensure `accent_primary` provides ≥ 3:1 contrast against `bg_primary`
- Table header: use the firm's dark brand color (e.g. deep navy / dark green) rather than
  a light value

### 0E — Confirm with the User (optional, brief)

After resolving tokens, output a **single short line** — not a table, not a list — before
generating the report:

> *"Using [firm_name] branding — [accent_primary] accent on [bg_primary] background,
> [font_heading] / [font_body] fonts."*

If brand detection produced low-confidence results (e.g. only a logo image was provided with
no color context), ask one focused question:

> *"I've picked up [X] and [Y] from your file — is that the right color scheme,
> or would you like to adjust anything?"*

Do not ask if defaults were applied — just proceed.

---

## Step 1 — Determine Report Mode

**Data source mode:**

| Signal | Mode |
|---|---|
| User uploads an Excel model (.xlsx) or references an uploaded file | **Model mode** — read data from the file |
| No file uploaded; company exists in the connected data platform | **MCP mode** — fetch live from the platform |
| Both available | Prefer the model for financials and commentary; use MCP to supplement metadata and returns |

**Report type:**

| User asks for… | Report type |
|---|---|
| One-pager, tearsheet, company report, GP report | **GP One-Pager** |
| LP update, quarterly update, LP quarterly report | **LP Quarterly Update** |

If unclear, default to **GP One-Pager**.

---

## Step 2 — Resolve the Company & Fetch Data

### MCP Mode

All data is fetched from the user's connected **Chronograph MCP server**. The agent must have the Chronograph MCP connected before running the skill in MCP mode; if it isn't connected, prompt the user to connect it, or fall back to Model mode if a file is available.

Inspect the connected Chronograph MCP's tool list at runtime and pick the appropriate tool for each step below based on the tool descriptions the server provides. Do not hard-code tool names — read the live descriptions to stay current.

### Fetch sequence

Work through these steps in order. At each step, identify the right Chronograph tool from its description, call it to fetch what the report needs, and hold the result for the rendering steps that follow. The skill does not need a fixed schema — only the values listed below in plain language.

**1. Resolve the company.** Find the tool that searches for companies, funds, and other portfolio entities by name. Use it to turn the user's input into a canonical company. If multiple candidates come back, prefer the closest name match; ask the user only if genuinely ambiguous.

**2. Get the company's basic facts.** Use the tool that retrieves core portfolio entity records. Fetch what the **Header Strip** (Step 4) needs: company name, sector, industry, geography, HQ, and the company's reporting currency. Request only what the header renders — the tool's description will list what's available.

**3. Get the company's investments.** Using the same core-entity retrieval tool, fetch the list of investments associated with the company. The **Investment Returns Table** (Step 4) needs one row per investment: investment name, the fund it belongs to, entry date, exit date if applicable, and whether it has exited.

**4. Get the company's financial line items.** Find the tool for company-level financial metrics. Make a discovery / help call first, passing the company ID, to see what's available.

**Metric selection rule.** Always query by the platform's metric type when one is available for the line item — even if the company's display label differs. Use a company-specific mapped metric definition only when no metric type covers the line item or the request is inherently tenant-specific (bespoke KPIs, commentary fields, operating metrics, custom valuation inputs).

Priority:

1. Platform metric type
2. Company-specific mapped metric definition (only if no metric type exists)
3. Name-based metric search (only if neither of the above resolves)

Do not pick a company-specific metric because its label is a closer word match than an available metric type. Metric types drive querying; display labels drive presentation.

What the report needs (query via the metric type whenever available):

- **Financial Performance Table:** Revenue, Gross Profit, EBITDA, the company's adjusted EBITDA series, and Net Debt. Pull the last 4–5 trailing-twelve-month periods plus the most recent quarter.
- **Valuation Summary Table:** Enterprise Value, valuation multiple, Total Debt, Cash, Net Debt, Equity Value — current quarter and prior quarter.
- **KPI Strip:** the latest values for LTM Revenue, LTM Adj. EBITDA, Enterprise Value, and Net Debt.

Use LTM for income-statement items, As-of for balance items. The discovery response will tell you which period types are supported per metric.

**5. Get per-investment returns.** Find the tool for investment-level performance. It uses gross performance figures (not net — net lives elsewhere on the platform and is the wrong tool for the GP one-pager). The **Investment Returns Table** needs, per investment: invested capital, realized proceeds, unrealized value, MOIC, and IRR. Sum across investments to populate the MOIC card in the **KPI Strip**.

**6. Resolve any non-standard metrics by name.** If the user asks the report to include a tenant-specific KPI that isn't part of the platform's standard metric set — e.g. a custom operating metric — use the metric-name search tool to resolve it to an ID before querying the company-metrics tool.

### Currency

Use the company's reporting currency as returned in step 2. **Do not default to USD.** If the user explicitly asks for a different currency, pass that through to each tool call.

### If something is missing

- Company not found → ask the user to confirm spelling; try the legal name.
- A metric returns no value → display `—`. Never fabricate.
- A tool returns a schema or help error → comply with the introspection call it asks for and retry.
- Chronograph MCP not connected → prompt the user to connect it, or fall back to Model mode.

### Model Mode

Read the uploaded Excel file using pandas (`data_only=True`). Extract from:

| Tab | Data to extract |
|---|---|
| `Overview` | Company name, HQ, fiscal year end, fund name(s), investment names, entry dates, ownership % |
| `Performance` | Revenue, Gross Profit, EBITDA, Adj. EBITDA (Reported and Valuation basis), Net Debt — last 4–5 LTM periods |
| `Valuation` | EV, valuation multiple, equity value, net debt — current and prior quarter; commentary text (Business Update, Rationale for Conclusion, Rationale for Discount); per-investment cost, realized, unrealized, MOIC |

**Currency and scale:** Check the `Figures In` field on the Overview tab. If `1000`
(thousands), divide all financial values by 1,000 before displaying in millions. Note the
`Local Currency` field and include it in the report header.

**If commentary is present in the model**, use it verbatim (lightly edited for grammar only).

---

## Step 3 — Layout Structure

Both report types share the same structural template. The Financial Performance table must
always be **directly above** the Valuation Summary table in the same column.

```
┌─────────────────────────────────────────────────────────────────┐
│  HEADER — Logo · Company · Fund · Sector · HQ · Report date    │
│           Tag pills: sector, geography, fund, status           │
├─────────────────────────────────────────────────────────────────┤
│  KPI STRIP (5 cards, full width)                               │
│  LTM Revenue | LTM EBITDA | Enterprise Value | Net Debt | MOIC │
├───────────────────────────┬─────────────────────────────────────┤
│  LEFT COLUMN              │  RIGHT COLUMN                       │
│                           │                                     │
│  Financial Performance    │  Revenue & EBITDA Bar Chart         │
│  Table                    │  (last 4 LTM periods, SVG inline)  │
│                           │                                     │
│  Valuation Summary        │  Investment Returns Table           │
│  Table  ← must stay here  │                                     │
│                           │                                     │
├───────────────────────────┴─────────────────────────────────────┤
│  COMMENTARY (full width, up to 3 columns)                      │
│  Business Update | Valuation Rationale | Discount Rationale    │
├─────────────────────────────────────────────────────────────────┤
│  FOOTER — © {firm_name} | {website} | {confidentiality_label} │
└─────────────────────────────────────────────────────────────────┘
```

---

## Step 4 — Panel Specifications

### Header Strip

- Background: `bg_header` → `bg_primary` gradient, 135°
- **Logo:** if `logo_url` is set, render `<img src="{logo_url}">` at `logo_position`;
  maintain clearspace equal to the cap-height of the firm name on all sides
- **Logo fallback:** if `logo_url` is blank, render firm name in `font_heading`,
  `font_heading_weight`, 24px, `text_primary`
- Company name: `font_heading`, 28px, `accent_primary`
- Subtitle (fund · sector · HQ · date): `font_body`, 12px, `text_muted`
- Tag pills: `bg_secondary` fill, `accent_secondary` border, `font_body` 10px

### KPI Strip (5 cards)

| Card | Value |
|---|---|
| LTM Revenue | Current period revenue |
| LTM Adj. EBITDA | Valuation basis if available, else Reported |
| Enterprise Value | Current quarter EV |
| Net Debt | Current quarter net debt |
| Gross MOIC | Blended or primary investment MOIC |

- Card: `bg_secondary` background, 6px border-radius, `box-shadow: 0 2px 8px rgba(0,0,0,0.4)`
- Value: `font_heading`, 28px, `accent_primary`
- Label: `font_body`, 9px, uppercase, letter-spacing 0.05em, `text_primary`
- Delta: `▲ +X%` in `accent_positive` · `▼ -X%` in `accent_negative`

### Financial Performance Table (LEFT column, top)

Rows (LTM, 3 prior periods + current quarter, vs. PY column):

| Row | Notes |
|---|---|
| Revenue | |
| Gross Profit | |
| Gross Margin % | |
| EBITDA | |
| EBITDA Margin % | |
| **Adj. EBITDA (Valuation)** | Highlight row — `font_heading`, `accent_primary` value |
| **Adj. EBITDA Margin %** | Highlight row |
| Net Debt | |

- Header row: `table_header_bg`, `font_body` 9px uppercase, `text_primary`
- Alternating rows: `bg_secondary` / `bg_primary`
- Highlight rows: `rgba({accent_primary}, 0.06)` background
- Numeric columns right-aligned; label column left-aligned

### Valuation Summary Table (LEFT column, directly below Financial Performance)

| Row | Notes |
|---|---|
| Valuation Multiple | `X.Xx` |
| Adj. EBITDA (LTM) | currency |
| **Enterprise Value** | Highlight row |
| Total Debt | |
| Cash | |
| Net Debt | |
| **Total Equity Value** | Highlight row |

Prior Quarter vs Current Quarter, plus a delta column (`accent_positive` / `accent_negative`).

### Revenue & EBITDA Bar Chart (RIGHT column, top)

- **Inline SVG only** — no external JS or D3
- Side-by-side bars: Revenue in `accent_secondary`, Adj. EBITDA in `accent_negative`
- Current quarter Revenue bar: `accent_primary`
- 4 LTM periods on x-axis; `$Xm` labels above each bar
- Current quarter label: `font_heading`, `accent_primary`; prior: `font_body`, `text_muted`
- Y-axis gridlines: dashed, `rgba({accent_primary}, 0.08)`
- Legend: colour swatches + labels, `font_body` 9px
- Auto-scale: y-axis max = largest revenue × 1.2; bar heights proportional

### Investment Returns Table (RIGHT column, bottom)

| Column | Format |
|---|---|
| Investment name | Left-aligned |
| Entry date | `Mon YYYY` |
| Ownership % | `X.X%` |
| Cost (Gross) | `$Xm` |
| Realized | `$Xm` |
| Unrealized | `$Xm` |
| MOIC (Gross) | `X.XXx` — `font_heading`, `accent_primary` |

- Sort: largest cost first
- **Total / Blended** row at bottom in `font_heading`
- Add IRR column if available; omit column if all values null
- Unavailable values: `—`
- One-sentence QoQ note below table: `font_body` 9.5px, `text_muted`

### Commentary Section (full width, up to 3 columns)

1. **Business Update** — verbatim from model if present; AI-generated in MCP mode
2. **Rationale for Valuation Conclusion** — verbatim from model if present
3. **Rationale for Discount to Comps** — verbatim from model if present

- Eyebrow label: `accent_primary`, 9px, uppercase, letter-spacing 0.08em,
  1px bottom border in `accent_primary`
- Body: `font_body`, 11px, `text_muted`, line-height 1.6
- Collapse empty columns — never show a blank block

---

## Step 5 — LP Quarterly Update Additions

When report type is **LP Quarterly Update**:

1. Header eyebrow → `LP QUARTERLY UPDATE · Q[N] YYYY`
2. Commentary must come from model verbatim — never AI-generate for LP reports
3. Add an optional **valuation bridge note** below the Valuation Summary table:
   *"EV increase driven by multiple expansion from X.Xx to Y.Yx; EBITDA contribution +$Zm"*

---

## Step 6 — CSS Variables & Font Loading

Populate these from the resolved brand tokens (Step 0). Set once in `<style>`; all panels
inherit automatically.

```html
<!-- In <head> -->
<link href="{google_fonts_url}" rel="stylesheet">

<style>
:root {
  --bg-primary:        {bg_primary};
  --bg-secondary:      {bg_secondary};
  --bg-header:         {bg_header};
  --accent-primary:    {accent_primary};
  --accent-secondary:  {accent_secondary};
  --accent-negative:   {accent_negative};
  --accent-positive:   {accent_positive};
  --text-primary:      {text_primary};
  --text-muted:        {text_muted};
  --table-header-bg:   {table_header_bg};
}

body {
  font-family: '{font_body}', sans-serif;
  font-weight: {font_body_weight};
  background: var(--bg-primary);
  color: var(--text-primary);
  margin: 0;
}

h1, h2, h3, h4, .eyebrow, .kpi-value, .moic-value {
  font-family: '{font_heading}', sans-serif;
  font-weight: {font_heading_weight};
}
</style>
```

---

## Step 7 — Formatting Rules

| Type | Format |
|---|---|
| Currency (millions) | `$43.5m` |
| Negative / net cash | `($3.4m)` in `var(--accent-negative)` |
| Multiples | `1.41x` |
| Percentages | `15.7%` |
| Basis point changes | `+70bps` |
| Dates | `Q4 2024` or `31 Dec 2024` |
| MOIC | `4.97x` — heading font, `var(--accent-primary)` |
| Positive delta | `▲ +X%` — `var(--accent-positive)` |
| Negative delta | `▼ -X%` — `var(--accent-negative)` |
| Unavailable | `—` (em dash) — never fabricate |

---

## Step 8 — Output

- **File name:** `[company_name_lowercase_underscored]_[report_type]_[quarter].html`
  - Examples: `ashworth_health_gp_report.html` · `ashworth_health_lp_update_q4_2024.html`
- **Single self-contained HTML file** — all CSS in `<style>`, all data inline
- **No external JS** — bar chart is inline SVG
- **Footer:** `© {firm_name} | {website} | {confidentiality_label}` (omit blank fields)
- Save to the outputs directory and present the file to the user

---

## Error Handling

| Situation | Action |
|---|---|
| No brand template and no defaults configured | Apply Chronograph defaults silently |
| Brand template provided but colors are ambiguous | Apply best-guess tokens; surface the one-line confirmation (Step 0E) |
| Logo URL returns 404 or is empty | Fall back to firm name as heading-font text |
| Company not found via MCP | Ask user to confirm spelling; try alternate names |
| Metric unavailable | Display `—`; never fabricate |
| Commentary empty in model | AI-generate; label `(AI-generated)` in small `text_muted` |
| IRR null for all investments | Omit IRR column |
| Cost split unavailable for a tranche | Show combined row; note gap in footnote |
| Model and MCP figures conflict | Flag in footnote; prefer model figures |
| `Figures In` = 1000 | Divide all values by 1,000 before displaying in millions |
| Light-theme brand detected | Swap text tokens to dark; verify contrast ≥ 4.5:1 before rendering |
| Chronograph MCP not connected (MCP mode) | Prompt user to connect; fall back to Model mode if a file is available |

---

## Data & Brand Checklist (verify before rendering)

**Brand**
- [ ] Brand tokens resolved — from uploaded template or Chronograph defaults
- [ ] One-line brand confirmation output to user (or low-confidence question asked)
- [ ] CSS variables set; Google Fonts `<link>` in `<head>`
- [ ] Logo rendered or firm-name fallback applied
- [ ] Theme (dark/light) confirmed; contrast ratios verified

**Data**
- [ ] Data source mode determined (Model or MCP)
- [ ] Report type determined (GP One-Pager or LP Quarterly Update)
- [ ] Currency and scale confirmed (`Figures In` handling applied if needed)
- [ ] Financial Performance table: 4 periods + vs. PY column
- [ ] Valuation Summary: PQ vs CQ + delta — in same column as Financial Performance
- [ ] Bar chart: 4 periods, bars labelled, y-axis auto-scaled
- [ ] Investment Returns: per-investment rows + total row
- [ ] Commentary: model text verbatim if available; empty blocks collapsed
- [ ] Footer: firm name, website, confidentiality label (omit blanks)
- [ ] Single self-contained HTML file, no external JS

Referenced files: 1