The problem

Financial narratives and investor decks are designed to sell, often obscuring fragility through aggressive add-backs, misleading cash conversion, or hidden leverage. Generic AI summaries fail to catch these fractures, resulting in poor underwriting and expensive investment mistakes.

What it does

• Performs forensic accounting reviews to verify if reported earnings are backed by actual cash flow.
• Stress tests financial models against aggressive assumptions to identify the most likely break points.
• Identifies fraud red flags, governance issues, and misaligned management incentives.
• Constructs detailed downside scenarios, including mild, severe, and total failure cases.
• Produces a diligence matrix ranking revenue quality, debt realism, and concentration risks.

Why this beats prompting it yourself

Most prompts generate summaries that echo management's optimism. This skill enforces a forensic skepticism layer that reconciles stories against balance sheet truth, tracing every claim to evidence while explicitly flagging what is missing. It moves beyond data extraction to provide decision-grade judgment on whether to pursue or walk away.

Use cases

• Conducting Quality of Earnings (QoE) reviews for M&A and private equity deals.
• Evaluating venture capital memos and startup fundraising decks for durability.
• Performing credit assessments and covenant headroom analysis for lenders.
• Auditing public company quarterly reports for non-GAAP anomalies and margin quality.

Known limitations

Requires specific financial data (statements, decks, or models) to function. If evidence is insufficient, the engine will halt rather than invent conclusions.