Risk Management Agent¶

Trade validation, position sizing, and portfolio limit enforcement

December 2025 Status

ACTIVE - Enforcing risk rules on all trades

Rule	Status	Value
Max Position Size	✅ Enforced	$1,000
Max Portfolio %	✅ Enforced	20% concentration
Min Cash Reserve	✅ Enforced	50%
Min Risk/Reward	✅ Enforced	2.5:1
Sentiment Blocking	✅ Active	4hr block on bearish alerts

Overview¶

The Risk Management Agent validates trades, calculates position sizes, and enforces portfolio limits.

trading_agents/risk_management_agent.py

Responsibilities: - Calculate position size based on risk parameters - Validate trades against portfolio limits - Enforce cash reserve requirements - Approve or reject trade requests - Handle sentiment alerts for position blocking

Risk Rules¶

Rule	Default	Description
Max Position Size	$1,000	Maximum value per position
Max Portfolio %	20%	Maximum position as % of portfolio
Min Cash Reserve	50%	Minimum cash reserve requirement
Min Risk/Reward	2.5:1	Minimum risk/reward ratio
Stop Loss	Required	All positions must have stops

Event Architecture¶

Events Published¶

Event	Description	Data
`risk.trade_approved`	Trade passed all risk checks	symbol, quantity, position_value
`risk.trade_rejected`	Trade violated risk rules	symbol, reason, checks_failed
`risk.size_calculated`	Position size determined	symbol, quantity, buying_power
`risk.review_position`	Review triggered by sentiment	symbol, recommendation

Events Subscribed¶

Event	Handler	Description
`strategy.entry_signal`	`handle_strategy_entry_signal`	Validate entry requests
`risk.calculate_size`	`handle_risk_calculate_size`	Calculate position size
`risk.check_limits`	`handle_risk_check_limits`	Report current limits
`sentiment.alert`	`handle_sentiment_alert`	Block bearish symbols

Risk Checks Performed¶

Trade Approval Flow¶

graph TD
    A[Entry Signal] --> B{Symbol Blocked?}
    B -->|Yes| C[REJECT: Bearish Sentiment]
    B -->|No| D{R/R Ratio OK?}
    D -->|No| E[REJECT: R/R Too Low]
    D -->|Yes| F{Position Size OK?}
    F -->|No| G[REJECT: Size Exceeded]
    F -->|Yes| H{Portfolio % OK?}
    H -->|No| I[REJECT: Concentration]
    H -->|Yes| J{Cash Reserve OK?}
    J -->|No| K[REJECT: Insufficient Cash]
    J -->|Yes| L{Buying Power OK?}
    L -->|No| M[REJECT: No Buying Power]
    L -->|Yes| N[APPROVE TRADE]

Check Details¶

Check	Validation	Failure Message
Symbol Blocked	Not in bearish sentiment list	`symbol_blocked_sentiment`
Risk/Reward	ratio >= 2.5:1	`risk_reward_too_low`
Position Size	value <= $1,000	`position_size_exceeded`
Portfolio Concentration	position/portfolio <= 20%	`portfolio_concentration_exceeded`
Cash Reserve	remaining cash >= 50%	`cash_reserve_insufficient`
Buying Power	position <= buying power	`buying_power_insufficient`

Sentiment Integration¶

Bearish Sentiment Blocking¶

When sentiment.alert with HIGH severity BEARISH is received:

Symbol added to blocked_symbols set
Block expires after 4 hours (configurable)
New entry signals for blocked symbols are rejected
risk.review_position published for open positions

Bullish Sentiment Handling¶

Symbol removed from blocked list
Position size multiplier applied (up to 1.25x)

Configuration¶

# Risk Management Agent defaults
self.sentiment_block_duration_hours = 4
self.sentiment_veto_threshold = -0.6
self.sentiment_boost_threshold = 0.5
self.sentiment_boost_multiplier = 1.25

Position Size Calculation¶

Algorithm¶

# Base calculation
adjusted_max_position = max_position_size * position_size_multiplier
position_value = min(adjusted_max_position, buying_power)
quantity = int(position_value / entry_price)

Factors Affecting Size¶

Factor	Impact
`max_position_size`	Hard cap at $1,000
`position_size_multiplier`	1.0-1.25x based on sentiment
`buying_power`	Available cash limits position
`entry_price`	Determines share quantity

Usage Examples¶

Initialize Agent¶

from trading_agents.risk_management_agent import RiskManagementAgent

agent = RiskManagementAgent(
    max_position_size=1000.0,    # $1,000 max
    max_portfolio_pct=0.20,      # 20% max concentration
    min_cash_reserve_pct=0.50,   # 50% cash reserve
    min_risk_reward=2.5,         # 2.5:1 minimum R/R
    paper=True                   # Paper trading mode
)

agent.start()

Check Risk Limits¶

# Request current limits
agent.publish_event(
    event="risk.check_limits",
    data={}
)

# Response:
# {
#     "max_position_size": 1000,
#     "max_portfolio_pct": 0.20,
#     "min_cash_reserve_pct": 0.50,
#     "buying_power": 66248.57,
#     "portfolio_value": 101305.76,
#     "cash": 66248.57,
#     "cash_reserve_pct": 0.654
# }

Calculate Position Size¶

agent.publish_event(
    event="risk.calculate_size",
    data={
        "symbol": "AAPL",
        "entry_price": 175.50
    }
)

# Response:
# {
#     "symbol": "AAPL",
#     "entry_price": 175.50,
#     "quantity": 5,
#     "position_value": 877.50,
#     "buying_power": 66248.57
# }

Circuit Breaker Protection¶

The agent uses circuit breakers for Alpaca API calls:

# Paper trading breaker
self.circuit_breaker = create_alpaca_paper_breaker()

# Protected API call
account = self.circuit_breaker.call(
    self.trading_client.get_account
)

Breaker States¶

State	Behavior
Closed	Normal operation
Open	Fail fast, skip API calls
Half-Open	Test single request

Logging Examples¶

Successful Approval¶

INFO Trade approved: AAPL | 5 shares @ $175.50 ($877.50)

Rejection¶

WARNING Trade rejected: AAPL | Reasons: cash_reserve_insufficient

Sentiment Block¶

WARNING RISK BLOCK: AAPL blocked for 4hr due to HIGH bearish sentiment (-0.72)

Configuration Reference¶

Parameter	Type	Default	Description
`max_position_size`	float	1000.0	Max position value in dollars
`max_portfolio_pct`	float	0.20	Max position as % of portfolio
`min_cash_reserve_pct`	float	0.50	Min cash reserve %
`min_risk_reward`	float	2.5	Min risk/reward ratio
`paper`	bool	True	Use paper trading account
`redis_host`	str	"localhost"	Redis host for message bus
`redis_port`	int	6379	Redis port