Derivatives risk exposure monitor · free AI skill

Produces a structured daily derivatives risk exposure report covering position-level exposure metrics, limit breach detection, key risk indicator (KRI) status, margin and liquidation risk, concentration analysis, stress…

Purpose

Convert a portfolio snapshot and a limit framework into a decision-ready risk surveillance report that surfaces every limit breach, flags KRIs trending toward breach, and ranks what the risk desk must act on today — replacing manual spreadsheet reconciliation with a repeatable daily read of exposure health.

When to Use

Invoke this skill when the user asks to:

Run the daily (or intraday) derivatives risk exposure report
Check open positions against notional, delta, VaR, or leverage limits
Identify which desks, books, or counterparties have breached or are near breach
Assess margin utilization and liquidation risk across the book
Evaluate concentration by underlying, expiry, counterparty, or venue
Produce a KRI dashboard status for a risk committee or MRM review

Inputs Required

Before starting, collect:

Portfolio positions: Position-level data — instrument, direction, notional, quantity, mark price, entry price (CSV, table, or structured list)
Limit framework: Hard and soft limits by metric and scope (e.g., notional cap per book, max net delta, VaR limit, leverage ceiling, single-underlying concentration %)
Greeks / risk sensitivities: Delta, gamma, vega, theta if available; otherwise directional exposure only
Margin data: Posted collateral, initial and maintenance margin, and current margin utilization per account or book
KRI definitions: The key risk indicators tracked, each with green/amber/red thresholds
Reporting scope: Whole firm, a specific desk, book, or counterparty
Stress scenarios: Shock assumptions to apply (e.g., ±20% underlying move, vol spike, funding rate spike) — optional

If greeks are unavailable, work from notional and directional exposure and flag the reduced fidelity. If no stress scenarios are supplied, apply a standard set: ±10% and ±20% underlying moves, and a +50% implied volatility shock.

Instructions

Step 1: Normalize and Aggregate Exposure

Build the exposure base before any comparison:

Compute mark-to-market value and unrealized P&L per position
Aggregate net and gross notional at each scope level (position → book → desk → firm)
Roll up net delta (directional exposure) and, where available, gamma, vega, theta
Compute leverage = gross notional ÷ equity (or ÷ posted collateral where equity is unavailable)
Note any positions with stale marks, missing greeks, or unmatched instruments as data-quality flags

Step 2: Limit Breach Detection

Compare every aggregated metric against its limit:

Classify each metric as OK (<80% of limit), Watch (80–99%), Soft Breach (≥100% of a soft limit), or Hard Breach (≥100% of a hard limit)
For each breach, report: scope, metric, current value, limit, % of limit, and the excess amount
Flag any breach that is both large in magnitude and fast-moving (utilization jump vs. prior period, if a prior snapshot is provided)
Do not soften language — a hard breach is reported as a hard breach

Step 3: Margin & Liquidation Risk

Assess how close the book is to a forced action:

Compute margin utilization = used margin ÷ available collateral per account/book
Flag Amber at ≥70% utilization and Red at ≥85%
Estimate the adverse price move (per major underlying) that would trigger maintenance margin call or liquidation
Identify concentrated liquidation risk: positions where a single move liquidates a disproportionate share of the book

Step 4: Concentration Analysis

Detect single-point-of-failure exposure:

Rank exposure by underlying, by expiry bucket, by counterparty, and by venue
Flag any single dimension exceeding its concentration limit (or 25% of gross exposure if no limit is defined)
Highlight expiry clustering (rollover/gamma risk) and counterparty concentration (credit risk)

Step 5: KRI Dashboard Status

For each supplied KRI, assign a status and a one-line read:

Map current value to green / amber / red per the supplied thresholds
State direction of travel (improving / stable / deteriorating) if a prior period is available
Call out any KRI newly crossing amber or red since the last run

Step 6: Stress Scenario Impact

Apply the scenarios (supplied or default) to the aggregated book:

Report projected P&L impact and post-shock margin utilization per scenario
Identify which scenario produces the first liquidation or hard-limit breach
Present as a compact scenario table: scenario → P&L impact → post-shock leverage → post-shock margin utilization → breach triggered (Y/N)

Step 7: Ranked Escalation List

Produce the action list, ordered by severity then speed of onset:

Hard breaches requiring immediate reduction or hedge
Margin/liquidation red flags
Soft breaches and concentration limit breaches
Watch-status metrics approaching limits
Data-quality gaps that must be resolved before the next run

For each item: what breached, by how much, the recommended action, and the owner/desk responsible.

Output Format

Structure: Executive summary banner first (breach count, worst KRI, highest-risk book), then sectioned body, then ranked escalation list
Length: 800–1,200 words
Format: Structured document with tables for breach detection, KRI status, and stress scenarios; numbered list for escalations
Sections: Summary banner → Exposure aggregation → Limit breaches → Margin & liquidation risk → Concentration → KRI dashboard → Stress scenarios → Ranked escalation list → Data-quality flags
Tone: Risk desk analyst — precise, quantitative, non-hedging; states breaches plainly and attaches numbers to every claim

Quality Checklist

Before delivering output, verify:

[ ] Every limit in the framework has been compared and assigned a status (OK / Watch / Soft Breach / Hard Breach)
[ ] Each reported breach includes current value, limit, % of limit, and excess amount
[ ] Margin utilization is computed per account/book with amber/red flags applied
[ ] Every supplied KRI appears in the dashboard with a green/amber/red status
[ ] The stress table shows at least three scenarios with P&L and post-shock margin impact
[ ] The escalation list is ordered by severity, not by appearance order, with an owner named per item
[ ] Data-quality gaps (stale marks, missing greeks) are explicitly flagged, not silently dropped

Notes

Fidelity scales with input quality: without greeks the report covers directional and notional risk only, and this limitation must be stated in the summary banner
Run against a prior snapshot whenever possible — direction of travel on utilization and KRIs is often more actionable than the point-in-time level
A "Watch" metric moving from 82% to 96% in one cycle is frequently more urgent than a stable soft breach; weight speed of onset in the escalation ranking
For crypto venues, funding rate and cross-margin contagion are common blind spots — include funding-driven P&L drift and cross-collateral liquidation cascades when the data supports it

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