private-equity

Use when the user wants a recorded video walkthrough of the Decision- Optimization Diagnostic (DX) upload UI — drop CSVs, watch the pipeline run, see the report rendered. Drives the local pe-mcp-web app via Playwright with caption overlays at each step, records to .webm, and prints a one-line ffmpeg command to convert to MP4. Reproducible, scripted, no manual screen capture required.

Use when a PE shop or portco needs to audit a deployed AI-agent fleet — inventory every registered agent, flag zombies (idle too long), runaway-cost agents (modeled monthly spend over a threshold), and misaligned agents (eval rubric fail), and produce a board-defendable pruning recommendation list with annual savings if the prunes land. Tackles the 40%-of-agentic- projects-cancelled-by-2027 risk Gartner is forecasting, and the agent-sprawl problem mega-funds (Vista, Thoma Bravo) face once they deploy AI agents at portco scale. Pure deterministic — no LLM call inside the tool, modeled telemetry stamped as modeled.

Use when an operating partner wants to surface cross-portco

Use when an LP has sent the GP an AI-section DDQ (Due-Diligence Questionnaire) — typically the new ILPA v2.0 (Q1 2026) AI governance / data / risk sections — and the GP needs a first-draft response packet built deterministically from the fund's existing AI-evidence artifacts in finance_output/, with every answer citing its source and a cross-answer consistency layer flagging contradictions before the LP does.

Use when a portco is post-close and the operating partner needs

Use when a PE professional needs an EU AI Act (Regulation 2024/1689)

Use when a PE professional needs a fast diligence pass over a

Use when you need to grade a PE document AI output (CIM extractor, DDQ generator, IC memo drafter, board memo) against its structured source-of-truth. Scores four dimensions — citation accuracy, hallucination rate, coverage, consistency — without an LLM call, so the grade is reproducible and defensible. Closes the trust gap that stops PE shops from putting AI output in front of an IC.

Use when a portco is preparing for exit (banker engagement, IM/CIM drafting, buyer-side AI diligence) and needs to pre-audit every $ of AI-attributable EBITDA it plans to claim. AlixPartners productized buyer-side AI diligence (the AI Disruption Score). Sellers have nothing. This skill is the seller-side twin — a defensible AI EBITDA proof pack with provenance ledger, methodology disclosure, sensitivity table, and defensibility checklist. The document a portco hands to its M&A advisor before banker engagement, so the buyer's AI diligence team finds nothing surprising.

Use when a PE professional needs to turn a Decision-Optimization Diagnostic OpportunityMap (or any structured decision recommendation with $ impact + cohort + counterfactual) into a board-defendable narrative memo. Closes the agency gap — the reference failure mode is the e-TeleQuote case where a model said "throttle 3 states" and management couldn't defend the why in the boardroom. This skill produces the language a managing director can read into a board meeting on Wednesday.

Use when an operating partner needs to fold N portcos' P&Ls (or loan tapes) — each in a different chart-of-accounts — into a single comparable view. Output is a unified CSV with frozen canonical columns, a mapping audit (per-cell provenance), a cohort-level anomaly digest (magnitude / sign / coverage), and an editorial-letterpress HTML report. Kills the monthly "normalize-by-hand-in-a-spreadsheet" ritual.

Use when a PE professional needs to assess which portfolio companies are ready for AI adoption, run a per-company AI readiness scoring with go/wait gate, identify and rank AI quick wins by EBITDA impact, build a phased AI implementation roadmap for a portfolio company, or assess AI adoption risks and change management barriers.

Use when a PE professional wants to benchmark portfolio companies against each other or track a single portco's diagnostic outcomes over time. Two modes — cross-portco (rank, archetype index, peer groups, LP-reportable HTML) and within-portco time-series (snapshot, trend, delta between dated snapshots). Consumes OpportunityMap JSON sidecars produced by dx_report. Claude-native, pandas-only.

Use when a PE professional needs a sector-tailored due diligence checklist, wants to track workstream completion, or needs to generate a data room request list. Covers Financial, Commercial, Legal, Tax, Technology, HR, and Regulatory workstreams with sector-specific additions for SaaS, Healthcare, Manufacturing, Consumer, and FinTech.

Use when a PE professional needs to prepare for due diligence meetings including management presentations, expert calls, and customer reference calls. Covers tailored question lists by function, red flag indicators, and follow-up tracking templates.

Use when a PE professional needs to screen a CIM or teaser against fund criteria, apply a pass/fail framework, or produce a one-page screening memo. Covers metric extraction, threshold testing, deal recommendation, and structured memo output.

Use when a PE professional needs to discover investment targets, profile CRM data, apply investment thesis criteria, score deal fit, or draft founder outreach templates. Covers sector filtering, revenue/size screening, and MCP-powered CRM CSV profiling.

Use when a PE professional needs to run a Decision-Optimization Diagnostic on a portfolio company — find the top repeated decisions being made badly (cross-section blind spots invisible in aggregate dashboards), quantify annual $ impact, and produce a ranked OpportunityMap plus board-ready memos. Reference pattern is the e-TeleQuote case ($9.7M/yr hidden in 3 state×source cells). Claude-native — Claude reasons over pandas aggregates, no sklearn.

Use when a PE professional needs to draft a structured Investment Committee memo, generate quantitative deal scoring via classify_investor, or produce an executive summary for IC presentation. Covers all 10 IC memo sections, returns analysis, value creation planning, and MCP-powered quantitative prospect scoring.

Use when a PE professional needs to train a regression model on portfolio data, predict liquidity risk scores for portfolio companies or acquisition targets, scan the full portfolio by risk tier, identify key risk drivers, or stress test under downside scenarios. Covers MCP-powered ML model training via liquidity_predictor and individual risk prediction via predict_liquidity.

Use when a PE professional needs to assess public market risk for portfolio companies or their comparable benchmarks, pull historical volatility analysis, retrieve Sharpe ratio and max drawdown metrics, run comprehensive stock analysis, compare risk profiles against market benchmarks, or produce a quarterly market risk report. Covers MCP-powered risk analysis via get_volatility, get_risk_metrics, and analyze_stock.

Use when a PE professional needs to run exploratory data analysis on CRM CSV exports, audit data completeness and fill rates, analyze deal pipeline distributions, detect outlier deals, assess data quality flags, or produce a pipeline health scorecard. Covers MCP-powered full EDA via ingest_csv with PE-specific interpretation.

Use when a PE professional needs to track portfolio KPIs, detect classification drift, generate quarterly dashboards, compare against market benchmarks, identify underperformers, or prepare board reporting packages. Covers traffic-light KPI frameworks, MCP-powered drift detection via classify_investor, and benchmark comparison via get_risk_metrics.

Use when a PE professional needs to train an ML classifier on CRM CSV exports, score individual investor prospects with confidence levels, batch-rank the deal pipeline, diagnose model performance, or retrain on fresh CRM data. Covers investor classification, deal pipeline scoring, and MCP-powered ML model training via investor_classifier and classify_investor.

Use when a PE professional needs to analyze public market comparables for valuation context, generate price comparison charts for public peer groups, produce correlation heatmaps to assess co-movement, identify appropriate comp sets for a PE target, or rank public peers by relative performance. Covers MCP-powered comp charts via compare_tickers and correlation analysis via correlation_map.

Use when a PE professional needs to calculate IRR/MOIC from entry and exit terms, run sensitivity analyses across entry multiple, exit multiple, and hold period, benchmark against public market comparables via get_returns and get_risk_metrics, model exit scenarios (IPO, trade sale, secondary), or attribute fund-level returns to individual deals.

Use when a PE professional needs to analyze revenue quality through ARR cohort waterfalls, compute LTV/CAC ratios and payback periods, measure net dollar retention with expansion/contraction/churn breakdown, assess revenue quality (recurring %, concentration, contract length), or profile cohort CSV data via ingest_csv before running cohort analysis.

Use when a PE professional needs to build an EBITDA bridge from current to target performance, create a structured 100-day post-acquisition plan, define KPI targets by function with quarterly milestones, identify and size revenue and cost improvement levers, or track the status of value creation initiatives against plan.