# Design System Shared formatting and styling reference for this skill. Skills that emit a styled HTML report use these conventions directly; the Excel model skill applies only the number formatting, analytical density, and citation rules. ## Number Formatting | Type | Format | Example | |------|--------|---------| | Currency (large) | `$X.Xbn` or `$X,XXXmm` | `$95.4bn`, `$2,345mm` | | Currency (small) | `$X.XX` or `$X,XXX` | `$6.08`, `$1,234` | | Percentages | One decimal + `%` suffix | `42.3%` | | Multiples | One decimal + `x` suffix | `8.5x EV/EBITDA` | | Growth rates | Signed + `%` suffix + context | `+12.3% YoY` | | Basis points | Signed + `bps` | `+150bps` | | Share counts | `X.XXbn` or `X,XXXmm` shares | `15.33bn shares` | Right-align all numbers in tables. Never display raw unformatted numbers. ## Analytical Density Every data point should include three layers where possible: 1. **The data point itself** — with Daloopa citation 2. **Context** — vs. prior period, vs. peers, vs. guidance, vs. consensus 3. **Implication** — what it suggests (margin expansion, deceleration, thesis risk) Example: > Revenue: [$95.4bn](https://daloopa.com/src/123) (+6.1% YoY, beat consensus by +2.3%) — acceleration from +4.8% last quarter driven by the flagship product cycle (illustrative, not current data) Avoid single-metric tables. Combine related metrics: Revenue + Revenue Growth + Gross Margin in one table. ## Table Conventions - **Columns** = time periods (left to right, chronological) - **Rows** = metrics (grouped by category: P&L, margins, per-share, balance sheet) - Include YoY growth rates as sub-rows in italics - Highlight beats/misses with notation: `$1.52 (beat +3.2%)` - Source citation row at bottom of every table: `Source: Daloopa (company filings)` - Group related metrics together — no single-metric tables ## Commentary Blocks After every major data table, include a 2-3 sentence interpretive commentary block: 1. **What the data shows** — trend, inflection, anomaly 2. **Why it matters** — competitive positioning, estimate revision risk, thesis confirmation/challenge 3. **What to watch** — upcoming catalyst, guidance change, peer divergence 4. **What could go wrong** — risks to the trend, bear case interpretation, sustainability concerns ## Analyst's Perspective All analysis is written from the perspective of a **long/short equity investor** conducting fundamental research. This means: **Be critical, not promotional.** Management narratives are marketing until proven by the data. When results look good, ask: Is this sustainable? Is it one-time? Is the market already pricing it in? When trends look bad, ask: Is this cyclical or structural? Is management acknowledging the problem or hiding behind adjusted metrics? **Challenge the numbers.** Look for quality-of-earnings issues: revenue pulled forward, unsustainable margin drivers (e.g., under-investing in R&D, favorable one-time items), channel stuffing signals, aggressive accounting changes. A beat isn't bullish if it's low-quality. **Be honest about valuation.** Don't anchor to the current stock price and work backwards. If a DCF implies 50% upside, say so — but also say what has to go right. If a stock trades at 40x earnings with decelerating growth, say that's expensive. The analysis should help the reader decide whether to own the stock, not confirm whatever they already believe. **Flag red flags explicitly.** When you see concerning patterns — deteriorating cash conversion, growing gap between GAAP and non-GAAP, rising DSOs, management selling stock, guidance cuts disguised as "conservatism," buybacks at all-time highs — call them out clearly. **Assign conviction.** Don't hide behind balanced language when the data points in a clear direction. If the bear case is more likely, say so. If growth is clearly decelerating, don't soften it with "remains resilient." Use precise language: "decelerating," "deteriorating," "unsustainable," "mispriced" when warranted. **Separate signal from noise.** Not every data point matters equally. Highlight what changes the thesis and deprioritize what doesn't. A 10bps margin fluctuation is noise; a 300bps margin decline is signal. ## Report Structure - **Header**: Company name, ticker, date, "Prepared by {FIRM_NAME}" - **Executive summary first**: 3-5 key takeaways, each one sentence - **Daloopa citations on every financial figure** — no uncited numbers - **Footer**: "Prepared by {FIRM_NAME} | Data sourced from Daloopa" on every report - Firm defaults to "Daloopa" if user doesn't specify a firm name; see `data-access.md` Section 4.5 - Section ordering follows the skill's defined structure ## Color Palette Consistent across charts and HTML: | Role | Color | Hex | |------|-------|-----| | Primary | Navy | `#1B2A4A` | | Secondary | Steel Blue | `#4A6FA5` | | Accent | Gold | `#C5A55A` | | Positive | Forest Green | `#27AE60` | | Negative | Crimson | `#C0392B` | | Light Gray | — | `#F8F9FA` | | Mid Gray | — | `#E9ECEF` | | Dark Gray | — | `#6C757D` | | Near Black | — | `#343A40` | Chart color sequence: Navy, Steel Blue, Gold, then grays. ## Chart Styling All charts follow these rules: - Grid lines: light gray (`#E9ECEF`), no box border - Labels: 11px, gray (`#6C757D`) - Title: 14px bold, navy - Source citation below every chart: `Source: Daloopa (company filings)` - Adjacent commentary block required for every chart ## Typography (Rendered Outputs — HTML) - Primary font: system sans-serif stack (Segoe UI, -apple-system, Arial) - Headers: bold, 18-28px, navy (`#1B2A4A`) - Body: 12-13px, dark gray (`#343A40`) - Table cells: 11px, tabular-nums for number alignment - Citations: 8-9px italic ## HTML Report Template The skill writes styled HTML directly using this template. Copy the full template (including all CSS) into every report. Write the analysis as HTML content inside the `` — use `

`, `

` for headings, `` for data tables, `
` for commentary blocks, `
` for prose, `
`/`
` for lists, and `` for Daloopa citation links. ```html
Prepared by {FIRM_NAME} | Data sourced from Daloopa
```

Where it stands