Purpose
Turn three raw financial statements into a coherent credit story — computing the ratios that matter, spotting where trends are inflecting, and separating genuine risk signals from noise — so a banker walks into a client review or credit committee already knowing where the questions are.
When to Use
Invoke this skill when the user asks to:
- Review a borrower's or prospect's financial statements
- Underwrite a new loan or assess a credit request
- Evaluate an annual credit renewal or covenant compliance
- Compare a client's financials year-over-year and explain the changes
- Prepare talking points or questions before a client financial review
- Flag risk or opportunity signals hiding in a set of financials
Inputs Required
Before starting, collect:
- Income statement: Revenue through net income, ideally 2–3 periods
- Balance sheet: Assets, liabilities, equity, ideally 2–3 periods
- Cash flow statement: Operating, investing, financing sections if available
- Business context: Industry, company size, and what the business does
- Credit request or purpose: What the analysis is supporting (new facility, renewal, covenant check)
- Existing debt / covenants: Current facilities, terms, and any covenant thresholds
- Statement quality: Audited / reviewed / compiled / internal — and fiscal year-end
If cash flow is missing, reconstruct approximate operating cash flow from net income plus non-cash items and working capital changes, and note it as an estimate. If only one period is provided, run the analysis but flag that trend signals are unavailable.
Instructions
Step 1: Statement Integrity & Quality Check
Before interpreting numbers, sanity-check them:
- Does the balance sheet balance (assets = liabilities + equity)?
- Does net income tie to retained earnings movement plus dividends?
- Does ending cash on the cash flow statement tie to the balance sheet?
- Note the statement tier: audited > reviewed > compiled > internal. Lower tiers require more skepticism and heavier reliance on trend logic.
- Flag any one-time items, related-party transactions, or accounting changes that distort comparability.
Step 2: Profitability & Margin Analysis
Compute and interpret across all available periods:
- Gross margin, operating margin, net margin — direction and magnitude of change
- Revenue growth rate — organic vs. driven by pricing, one-time events, or acquisition
- EBITDA and EBITDA margin — the primary earnings proxy for debt service
- Diagnose: Is margin compression a cost problem, pricing problem, or mix problem? Is revenue growth profitable or is the company buying growth?
Step 3: Liquidity & Working Capital Analysis
- Current ratio and quick ratio — with context; a low ratio isn't automatically bad in some industries
- Working capital trend in absolute dollars
- Days Sales Outstanding, Days Inventory Outstanding, Days Payable Outstanding and the resulting cash conversion cycle
- Diagnose: Is the company financing growth by stretching payables or building receivables it can't collect? A lengthening cash conversion cycle is an early distress signal.
Step 4: Leverage & Solvency Analysis
- Debt-to-equity and debt-to-EBITDA (or Total Funded Debt / EBITDA) — the headline leverage read
- Debt Service Coverage Ratio (DSCR): (EBITDA − cash taxes − unfinanced capex) / (annual principal + interest) — the single most important credit ratio
- Interest coverage (EBIT / interest expense)
- Tangible net worth — equity less intangibles and goodwill
- Diagnose: Can the business service the requested debt from operating cash flow, or does repayment depend on refinancing or asset sales?
Step 5: Cash Flow Quality Analysis
- Operating cash flow vs. net income — persistent OCF below net income signals earnings quality issues
- Free cash flow (OCF − capex) — is the business self-funding?
- Financing section read — is the company drawing debt or raising equity to cover operating shortfalls?
- Diagnose: Profit without cash is the classic path to failure. Confirm reported earnings are converting to cash.
Step 6: Risk Signal Scan
Flag each of the following that is present, with the supporting number:
- Declining margins for 2+ consecutive periods
- DSCR below 1.25x
- Debt-to-EBITDA above industry norm or rising sharply
- Cash conversion cycle lengthening materially
- OCF negative or persistently below net income
- Rising receivables outpacing revenue (potential collection or channel-stuffing issue)
- Covenant thresholds approached or breached
- Negative or eroding tangible net worth
Step 7: Opportunity Signal Scan
Flag positive signals a relationship manager can act on:
- Consistent margin expansion or improving DSCR (capacity for more credit)
- Strong free cash flow (deposit, treasury, or investment product opportunity)
- Falling leverage (candidate for repricing or expanded facility)
- Working capital efficiency gains (line utilization or cash management fit)
Step 8: Credit Read & Recommended Questions
- Deliver a one-paragraph creditworthiness read: overall direction, primary risk, and primary strength.
- Assign a directional risk posture: Strengthening / Stable / Watch / Deteriorating — with the two numbers that drive the call.
- List 5–7 specific questions to ask the client, each tied to a number in the analysis (e.g., "Receivables grew 40% while revenue grew 12% — what changed in collections?").
Output Format
- Structure: Ratio tables per section followed by a short interpretation paragraph; risk and opportunity signals as flagged bullet lists; closing credit read and question list
- Length: 900–1,300 words
- Format: Structured document with tables for computed ratios and bullet lists for signals
- Sections: Integrity check → Profitability → Liquidity → Leverage & DSCR → Cash flow quality → Risk signals → Opportunity signals → Credit read + questions
- Tone: Credit analyst — precise, numerically grounded, skeptical without being alarmist; every conclusion cites a figure
Quality Checklist
Before delivering output, verify:
- [ ] DSCR, debt-to-EBITDA, and current ratio are all computed and shown with their inputs
- [ ] Every risk and opportunity signal cites the specific number that triggered it
- [ ] Trends are read across all provided periods — not a single-period snapshot when multiples exist
- [ ] Any reconstructed or estimated figure (e.g., inferred cash flow) is explicitly labeled as an estimate
- [ ] The credit read assigns one of the four risk postures, not a hedge
- [ ] Recommended client questions each map to a figure in the analysis
Notes
- DSCR is the most decision-relevant ratio for lending; when data is thin, prioritize getting it right over computing every secondary ratio.
- Profit without cash conversion is the recurring early warning — always cross-check net income against operating cash flow.
- Ratio benchmarks vary widely by industry; interpret against sector norms rather than fixed universal thresholds, and say so when a "weak" ratio is actually normal for the industry.
- Lower statement tiers (compiled, internal) warrant more weight on trend logic and less on any single reported figure.
How to use this skill
This file is a Claude Code skill in Anthropic's standard format — markdown with YAML frontmatter at the top.
Installing in Claude Code:
~/.claude/skills/<folder-name>/SKILL.md— auto-loads in every session anywhere on your machine (recommended for skills you'll use across multiple projects)<your-project>/.claude/skills/<folder-name>/SKILL.md— auto-loads only when working in that project (recommended for project-specific skills)
Restart Claude Code or open a new session to register.
Using in Claude.ai, ChatGPT, Gemini, or other AI tools:
- Drag this file into a new chat, or copy and paste its contents
- Send your task or request as your next message — the AI will apply the skill where it fits
For Claude Code skill details: https://docs.anthropic.com/en/docs/claude-code/skills