Financial Report

Financial Report

Overview

Use this skill to transform completed analysis into a polished financial report with a clear investment narrative, publication-ready structure, and supporting charts. Focus on editorial packaging, visual clarity, and analytical communication rather than file conversion alone.

Quick Start

• Start only after the core investment analysis or dataset already exists.
• Identify the asset or topic, target audience, report date, author or desk if provided, and the intended stance or key takeaway.
• Decide whether the report is recommendation-led, market-update-led, valuation-led, or data-dashboard-led before writing.
• Read references/financial-report-template.md for the canonical report skeleton.
• Read references/style-rules.md for tone and section ordering.
• Read references/chart-playbook.md when the report includes indicators, OHLCV, line, pie, bar, or comparison charts.
• Read references/chart-rendering-backends.md when the report must generate chart images for downstream documents.

Core Workflow

1. Build the report frame

• Create a title that combines the subject, the main hook, and the current stance or focus.
• Add a compact metadata block with date, coverage target, time horizon, and recommendation when available.
• Write an executive summary that makes the report's main conclusion clear immediately.

2. Lead with the investment view

• Surface the recommendation, thesis, valuation stance, or market view before background detail.
• Present the main reasons to own, avoid, monitor, or rotate into the asset in concise bullets.
• Keep the report directional, not merely descriptive.

3. Organize the body like professional finance research

• Use a consistent order: summary, thesis, subject overview, key drivers, valuation or metric context, charts and exhibits, catalysts, risks, and conclusion.
• Convert raw notes into claim-evidence form rather than source-dump prose.
• Keep tables and charts close to the argument they support.

4. Add charts intentionally

• Choose chart types based on the analytical question rather than decoration.
• Use OHLCV or candlestick-style charts for price action and market structure.
• Use line charts for time-series comparisons, indicators, trends, and benchmark-relative moves.
• Use bar charts for ranked comparisons, period-over-period changes, and peer snapshots.
• Use pie charts only for composition views such as revenue mix, portfolio weights, or ownership mix.
• Label axes, units, periods, legends, and sources clearly.
• Prefer TradingView lightweight-charts first for finance-native time-series visuals.
• Fall back to matplotlib.pyplot when the chart type is unsupported, awkward, or faster to render there.
• Export every chart as an image artifact that a downstream document generator can embed directly.

5. Adapt by asset type

• For equities, emphasize business model, earnings path, valuation range, catalysts, and operating charts.
• For crypto, emphasize protocol utility, adoption, token mechanics, valuation framing, on-chain or market indicators, and risk.
• For credit, emphasize cash generation, leverage, refinancing, spread behavior, covenant risk, and downside protection.
• For commodities, emphasize supply-demand setup, cycle position, macro drivers, inventory or curve data, and scenario ranges.
• For macro or portfolio reports, emphasize regime drivers, allocation logic, exposures, and benchmarking visuals.

6. Prepare for downstream rendering

• Output clean sectioned markdown or another neutral structured form unless the user explicitly asks for something else.
• Keep headings, tables, chart captions, footnotes, and disclaimers easy for an external formatting skill to preserve.
• Do not hard-code page-layout assumptions that belong to downstream docx or pdf tooling.

Output Requirements

• Include: title, summary, thesis or main view, supporting analysis, charts or exhibit list, risks, and conclusion.
• Keep the writing concise, directional, and publication-ready.
• Separate observed facts from judgment calls.
• Make charts analytical, not ornamental.
• Make sure the structure can be passed directly into an external formatting skill.

Guardrails

• Do not invent a recommendation unsupported by the source analysis.
• Do not add charts that do not answer a real question.
• Do not use pie charts for time-series data or too many categories.
• Do not let visual polish overwhelm analytical clarity.
• Do not rely on a specific file extension or rendering engine.
• Do not keep charts only as interactive widgets when the report requires embeddable document assets.