Stats
Actions
Tags
From fl03-skills
Research-grade quantitative finance. Loads QUANT.md (mathematical toolkit: stochastic calculus, derivative pricing, risk metrics, portfolio theory, numerical methods) and MODELS.md (applied model library: Black-Scholes, Heston, Merton jump-diffusion, term structure, factor models, Monte Carlo, backtesting). Imbues any agent with the working knowledge of a PhD-level financial engineer. Attach when the user asks about option pricing, the Greeks, implied vs realized volatility, Itô / SDEs / GBM, Value-at-Risk, CVaR / expected shortfall, Sharpe / Sortino / Calmar, Kelly sizing, mean-variance optimization, factor models (CAPM / Fama-French / APT), term-structure models (Vasicek / CIR / Nelson-Siegel), Monte Carlo or PDE-based pricing, or backtesting methodology. Analysis-only — pair with `@trader` (and an exchange skill such as `@polymarket`) for live execution.
How this skill is triggered — by the user, by Claude, or both
Slash command
/fl03-skills:financeThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
This skill is a primer + index. Depth lives in the two reference modules below. Read both
This skill is a primer + index. Depth lives in the two reference modules below. Read both when the task involves any quantitative finance work — they are designed to be loaded together.
| Module | Content |
|---|---|
QUANT.md | Core mathematical toolkit — stochastic calculus, probability, linear algebra, optimization, risk theory, market microstructure |
MODELS.md | Applied model library — Black-Scholes (PDE + closed form + Greeks), Heston, Merton jump-diffusion, term structure, factor models, Monte Carlo patterns, backtesting |
| Agent | Trigger | Job |
|---|---|---|
analyst | @finance:analyze {input} | Research-grade quantitative analysis — dispatches input through QUANT.md + MODELS.md, produces structured output (Assumptions / Methodology / Results / Caveats) |
See finance/agents/analyst.md for the full agent brief and dispatch template.
This skill is analysis-only. It contains no trading execution logic.
For live market trading, load @trader and an exchange-specific skill (e.g., @polymarket).
A quant does not predict. A quant prices risk. Every output is a distribution with a mean, variance, and tail. Point estimates without confidence intervals are not analysis — they are guesses.
The mathematical content in QUANT.md and MODELS.md is grounded in the standard quant
finance literature. When deeper rigor is needed, cite the canonical sources rather than
re-deriving from memory:
| Topic | Canonical source |
|---|---|
| Derivatives, Greeks, Black-Scholes intuition | Hull, Options, Futures, and Other Derivatives |
| Stochastic calculus, martingale pricing | Bjork, Arbitrage Theory in Continuous Time; Shreve, Stochastic Calculus for Finance II |
| Practitioner-flavored pricing & vol surface | Wilmott, Paul Wilmott on Quantitative Finance |
| Monte Carlo methods | Glasserman, Monte Carlo Methods in Financial Engineering |
| Portfolio theory, factor models | Grinold & Kahn, Active Portfolio Management; Cochrane, Asset Pricing |
Guides creation, editing, and verification of skills for AI coding agents using test-driven development with subagent scenarios. Use when authoring or debugging skills.
npx claudepluginhub fl03/claude --plugin webassembly