How do I install the Trading Risk Management skill?

Run npx skillmds@latest add knuckles-team/trading-risk-management in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Trading Risk Management safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Trading Risk Management?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Trading Risk Management free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Trading Risk Management?

Knuckles-Team (@knuckles-team) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Trading Risk Management by knuckles-team

Name: Trading Risk Management
Author: Knuckles-Team

File contents

Risk Management Skill-Graph — CONCEPT:AU-AHE.assimilation.skill-workflow-ingest / AU-OS.safety.prompt-injection-scanner

Comprehensive risk management patterns enforced by the Emerald Exchange RiskGuard engine.

P0 Risk Controls

Circuit Breakers

Control	Threshold	Action	Recovery
Portfolio Drawdown	> 10% from peak	HALT all trading	Manual resume only
Daily Loss Limit	> 3% of equity	HALT new orders	Next trading day
Regime Shift	KS-test p < 0.05	HALT + alert	Manual review
Kill Switch	Manual trigger	HALT immediately	`emerald_orders(action='resume')`

Position Sizing — Kelly Criterion

Kelly fraction = W - (1-W)/R
Where:
  W = historical win rate
  R = win/loss ratio (avg win / avg loss)

Capped at: min(kelly, max_position_pct=2%)

Pre-Trade Validation

Every order passes through RiskGuard.pre_trade_check():

Halt check: Is trading halted?
Position size: Does this exceed 2% equity limit?
Concentration: Portfolio exposure to single asset?
Cash availability: Sufficient buying power?
Live gate: If live mode, require human approval (OS-5.1)

Regime Detection — CONCEPT:EX-AHE.harness.ee-11

Hidden Markov Model (HMM)

3-state model: Bull, Bear, Neutral
Trained on rolling 252-day windows
State transitions trigger portfolio adjustments

KS-Test Regime Shift

Kolmogorov-Smirnov test on recent vs historical prediction distributions
Detects distribution drift before traditional drawdown signals
p-value < 0.05 triggers circuit breaker

Monitoring Schedule

Check	Interval	MCP Tool
Drawdown	5 minutes	`emerald_risk(action='drawdown_check')`
Daily P&L	5 minutes	`emerald_risk(action='daily_loss_check')`
Regime	15 minutes	`emerald_signals(action='regime')`
Portfolio snapshot	15 minutes	`emerald_portfolio(action='positions')`

Human-in-the-Loop — OS-5.1

Paper → Live promotion: Requires explicit human approval via graph_orchestrate.request_approval
Kill switch: Any agent can halt; only human can resume live trading
Audit trail: All risk events persisted as RiskSnapshot nodes in KG

Integration Points

Risk MCP: emerald_risk — status, drawdown, daily loss, Kelly, limits
Orders MCP: emerald_orders — halt/resume kill switch
KG Ontology: RiskSnapshot, regimeState, isHalted OWL classes
Workflow: risk-monitoring cron (every 5 min)

---name: trading-risk-managementdescription: Risk management patterns, circuit breakers, and compliance controls for the trading ecosystem.---# Risk Management Skill-Graph — CONCEPT:AU-AHE.assimilation.skill-workflow-ingest / AU-OS.safety.prompt-injection-scannerComprehensive risk management patterns enforced by the Emerald Exchange RiskGuard engine.## P0 Risk Controls### Circuit Breakers| Control | Threshold | Action | Recovery ||---------|-----------|--------|----------|| Portfolio Drawdown | > 10% from peak | HALT all trading | Manual resume only || Daily Loss Limit | > 3% of equity | HALT new orders | Next trading day || Regime Shift | KS-test p < 0.05 | HALT + alert | Manual review || Kill Switch | Manual trigger | HALT immediately | `emerald_orders(action='resume')` |### Position Sizing — Kelly Criterion```Kelly fraction = W - (1-W)/RWhere: W = historical win rate R = win/loss ratio (avg win / avg loss)Capped at: min(kelly, max_position_pct=2%)```### Pre-Trade ValidationEvery order passes through `RiskGuard.pre_trade_check()`:1. **Halt check**: Is trading halted?2. **Position size**: Does this exceed 2% equity limit?3. **Concentration**: Portfolio exposure to single asset?4. **Cash availability**: Sufficient buying power?5. **Live gate**: If live mode, require human approval (OS-5.1)## Regime Detection — CONCEPT:EX-AHE.harness.ee-11### Hidden Markov Model (HMM)- 3-state model: Bull, Bear, Neutral- Trained on rolling 252-day windows- State transitions trigger portfolio adjustments### KS-Test Regime Shift- Kolmogorov-Smirnov test on recent vs historical prediction distributions- Detects distribution drift before traditional drawdown signals- p-value < 0.05 triggers circuit breaker## Monitoring Schedule| Check | Interval | MCP Tool ||-------|----------|----------|| Drawdown | 5 minutes | `emerald_risk(action='drawdown_check')` || Daily P&L | 5 minutes | `emerald_risk(action='daily_loss_check')` || Regime | 15 minutes | `emerald_signals(action='regime')` || Portfolio snapshot | 15 minutes | `emerald_portfolio(action='positions')` |## Human-in-the-Loop — OS-5.1- **Paper → Live promotion**: Requires explicit human approval via `graph_orchestrate.request_approval`- **Kill switch**: Any agent can halt; only human can resume live trading- **Audit trail**: All risk events persisted as `RiskSnapshot` nodes in KG## Integration Points- **Risk MCP**: `emerald_risk` — status, drawdown, daily loss, Kelly, limits- **Orders MCP**: `emerald_orders` — halt/resume kill switch- **KG Ontology**: `RiskSnapshot`, `regimeState`, `isHalted` OWL classes- **Workflow**: `risk-monitoring` cron (every 5 min)

Frequently asked questions

Trading Risk Management