From private-credit
This skill fires automatically when extracting financial data from borrower quarterly packages, compliance certificates, or any borrower-provided financial reporting for the purpose of updating a credit model.
npx claudepluginhub kd-alt20/private-credit-pluginThis skill uses the workspace's default tool permissions.
This skill fires automatically when extracting financial data from borrower quarterly packages, compliance certificates, or any borrower-provided financial reporting for the purpose of updating a credit model.
Generates design tokens/docs from CSS/Tailwind/styled-components codebases, audits visual consistency across 10 dimensions, detects AI slop in UI.
Records polished WebM UI demo videos of web apps using Playwright with cursor overlay, natural pacing, and three-phase scripting. Activates for demo, walkthrough, screen recording, or tutorial requests.
Delivers idiomatic Kotlin patterns for null safety, immutability, sealed classes, coroutines, Flows, extensions, DSL builders, and Gradle DSL. Use when writing, reviewing, refactoring, or designing Kotlin code.
This skill fires automatically when extracting financial data from borrower quarterly packages, compliance certificates, or any borrower-provided financial reporting for the purpose of updating a credit model.
Extract financial data top-down, following the natural order of the financial statements:
Start at the top of the P&L and work down:
Extract whatever the borrower provides. If they report only Revenue → Gross Profit → Adjusted EBITDA, extract those three lines and note the gaps.
Pay attention to how debt is reported — net of OID on the balance sheet vs. face value on the compliance certificate. Note any lease obligations, sale-leaseback financing, or other off-balance-sheet items that may count as debt for covenant purposes.
The workflow is identical regardless of whether the source is Excel or PDF:
Even when the borrower provides an Excel file, it is safer and cleaner to type the numbers manually rather than copy-paste. Copy-paste can introduce hidden formatting issues, linked references to the source file, or misaligned rows. Manually entering the numbers forces the analyst to look at each line item and consider how it maps to the model.
Borrower financials almost never use the same labels as your credit model template. The correct approach is to customize the template to match the borrower's reporting, not to force the borrower's data into rigid template labels.
When building a quarterly model for a new borrower for the first time:
After the first quarter is set up:
When a borrower introduces a new line item or label you have not seen before:
The compliance certificate is a signed PDF document, executed by the CFO or another company officer. It is the authoritative source for covenant calculations. The Excel financial package is less formal.
When numbers in the financial package differ from the compliance certificate, this almost always reflects a reporting difference, not an error:
The analyst's job: Identify the discrepancy, attempt to bridge it using the credit agreement definitions, and if the bridge is not possible with available information, raise the question with the management team or agent. Do not assume the cert is wrong. Do not assume the financials are wrong. Understand why the numbers differ.
An analyst should almost always be able to reconcile the debt number between the balance sheet, the compliance certificate, and the credit agreement terms. The ingredients are: balance sheet debt, OID adjustment, lease/financing treatment per the agreement, and letter of credit inclusion. If you cannot bridge the debt, something is missing — escalate.
The ability to reconcile EBITDA depends entirely on the quality of the borrower's reporting. With detailed reporting (full P&L down to net income, itemized adjustments), you can bridge. With minimal reporting (Revenue → Gross Profit → Adjusted EBITDA), you cannot — flag the limitation and work with what is available.
The most common issue is not data errors but insufficient data. Borrowers report the bare minimum required by the credit agreement:
Analyst response: Assess what you can build with the data provided. Identify what additional information would be needed for a complete credit assessment. Frame questions to management or the agent to fill the gaps.
For acquisition-active borrowers, a common trap is that acquired company financials are folded into the reporting without restatement of prior periods. This means:
Analyst response:
When borrowers report monthly: