skill-creator
Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an…
Activate for: FRTB, Fundamental Review of the Trading Book, market risk RWA, trading book capital, SA-TB, standardised approach trading book, sensitivities- based method, SBM, delta, vega, curvature, default risk charge, DRC, residual risk add-on, RRAO, internal models approach,
$ npx -y skills add panaversity/agentfactory-business-plugins --skill basel-rwa-market --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/basel-rwa-marketContext preview
The summary Claude sees to decide when to auto-load this skill.
Activate for: FRTB, Fundamental Review of the Trading Book, market risk RWA, trading book capital, SA-TB, standardised approach trading book, sensitivities- based method, SBM, delta, vega, curvature, default risk charge, DRC, residual risk add-on, RRAO, internal models approach,
name: basel-rwa-market description: > Activate for: FRTB, Fundamental Review of the Trading Book, market risk RWA, trading book capital, SA-TB, standardised approach trading book, sensitivities- based method, SBM, delta, vega, curvature, default risk charge, DRC, residual risk add-on, RRAO, internal models approach, IMA, expected shortfall, ES, P&L attribution, PLA, backtesting, GIRR, CSR, equity risk, commodity risk, FX risk, market risk capital, trading desk. NOT for: credit risk RWA under standardised or IRB approach (use basel-rwa-credit), capital adequacy ratios and buffer calculations (use basel-capital), IFRS 9 ECL provisioning (use ifrs9-ecl). metadata: version: "1.0" author: "Panaversity — The AI Agent Factory" standard: "Basel III FRTB (BCBS d352/d457) — jurisdiction implementations vary (load overlay)"
Market risk capital under FRTB replaces the pre-2016 VaR-based framework with a risk-sensitive approach that distinguishes between the Standardised Approach (SA) and the Internal Models Approach (IMA). The SA is the default and fallback; IMA requires desk-level regulatory approval including P&L attribution and backtesting.
FRTB applies to all instruments in the regulatory trading book. The trading book boundary is defined by strict criteria (intent to trade, ability to hedge, daily fair-value accounting) -- not by the bank's internal designation alone.
---
Market Risk Capital (SA) = SBM + DRC + RRAO
The SBM captures delta, vega, and curvature risks across seven risk classes.
| # | Risk Class | Key Risk Factors | | --- | --------------------------------- | ------------------------------------------------------------------------------------------- | | 1 | GIRR (General Interest Rate Risk) | Risk-free rates by tenor (3M, 6M, 1Y, 2Y, 3Y, 5Y, 10Y, 15Y, 20Y, 30Y), cross-currency basis | | 2 | CSR non-securitisation | Credit spreads by issuer, sector, tenor | | 3 | CSR securitisation (CTP) | Correlation trading portfolio credit spreads | | 4 | CSR securitisation (non-CTP) | Non-correlation trading securitisation spreads | | 5 | Equity | Equity spot prices, repo rates | | 6 | Commodity | Commodity prices by type (energy, metals, agricultural, other) and tenor | | 7 | FX | Exchange rates against reporting currency |
Step 1 -- Net Sensitivities: Calculate net delta, vega, and curvature sensitivities per risk factor within each risk class.
Step 2 -- Weighted Sensitivities: Multiply each net sensitivity by the prescribed risk weight for that risk factor. Delta risk weights vary by risk class and tenor (e.g., GIRR: 1.5%-2.4% by tenor; Equity: 15%-70% by market cap and economy; FX: 15%). Vega risk weights: uniform within risk class, calibrated to stressed implied volatility.
Step 3 -- Aggregation Within Bucket: Aggregate weighted sensitivities within each bucket using prescribed intra-bucket correlations (rho). K*b = sqrt( sum_i sum_j rho_ij * WS*i * WS_j ) where WS = weighted sensitivity, rho = intra-bucket correlation
Step 4 -- Aggregation Across Buckets: Aggregate across buckets using prescribed inter-bucket correlations (gamma). Capital = sqrt( sum*b sum_c gamma_bc * S*b * S_c ) where S_b = net bucket-level capital, gamma = inter-bucket correlation
Step 5 -- Three Correlation Scenarios: Calculate SBM under three correlation scenarios: (a) Medium correlations (prescribed rho and gamma) (b) High correlations (rho _ 1.25, gamma _ 1.25, capped at 1) (c) Low correlations (rho _ 0.75, gamma _ 0.75) SBM = MAX(SBM_medium, SBM_high, SBM_low)
Curvature risk captures the non-linear risk that delta sensitivities miss. Calculate by shocking each risk factor up and down by a prescribed amount, revaluing the portfolio, and computing the curvature charge: CVR*k = -min( V(x_k^up) - V(x_k) - s_k * delta*k, V(x_k^down) - V(x_k) + s_k * delta_k, 0 ) where s_k = prescribed shock size for risk factor k.
DRC captures jump-to-default risk for credit and equity instruments in the trading book -- the risk of an issuer defaulting between now and the next rebalancing period.
DRC Calculation: DRC = sum*i ( LGD_i * Notional*i * RW_i ) after hedging benefit Risk weights by credit quality: AAA 0.5%, AA 2%, A 3%, BBB 6%, BB 15%, B 30%, CCC 50%, Default 100% Hedging benefit: long-short offset within same issuer and seniority; partial offset across issuers in the same sector using prescribed hedge benefit ratios.
DRC is additive to SBM -- no diversification benefit between DRC and SBM.
RRAO is a simple gross notional charge for instruments with exotic risks not captured by SBM or DRC:
Instruments subject to RRAO include: barrier options, digital options, Asian options, basket options with non-standard payoffs, variance/volatility swaps, correlation products, weather derivatives.
RRAO is additive -- no diversification with SBM or DRC.
---
IMA is approved on a desk-by-desk basis by the regulator. Any desk failing IMA eligibility reverts to SA
🚀 Marketplace of domain-specific plugins for building enterprise AI agents. Enable AI agents to perform finance, banking, legal, and sales workflows using modular domain plugins.
Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an…
Activate for: load context, inject context, cross-domain context, context for this endeavour, relevant background, multi-domain context, what context do I…
Activate for: delegate, delegation record, handoff, delegation brief, follow-up, delegation quality, assign to, hand off to, delegation tracking, who should…
Activate for: daily digest, morning briefing, what's happening today, start of day brief, week ahead, Monday brief, weekly digest, start of week, what do I…
Activate for: situation brief, bring me up to speed, prep me for, pre-meeting brief, decision brief, context brief, catch me up, quick brief, what do I need to…
Activate for: meeting prep, meeting synthesis, before the meeting, after the meeting, D/A/F/Q/R, meeting brief, stakeholder prep, meeting notes synthesis,…