This server wraps System R's risk management API into MCP tools for AI trading agents. You get 68 operations spanning pre-trade validation, position sizing via G-formula, Monte Carlo simulation, drawdown analysis, regime detection, and performance scoring. The pre_trade_gate tool is the headline: pass it a symbol, direction, entry, stop, and equity, and it returns whether the trade passes Iron Fist validation plus exact share count and dollar risk. All the core risk tools are free. LLM-powered chat costs credits. It connects to 25 brokers including IBKR, Schwab, Binance, and Alpaca. Reach for this when you're building an agent that needs institutional risk checks before executing trades, not just market data and order routing.
Python SDK for the System R AI API Toolkit.
System R AI is a decision intelligence system for trading and investing. The API Toolkit gives developers access to finance tools for agents, Python workflows, notebooks, and backend services.
The SDK is designed for structured decision-support workflows: position sizing, risk checks, performance diagnostics, market structure analysis, journal records, memory search, and tool discovery.
System R is software for decision support. It is not financial advice, not a broker, not a signal service, and not a guarantee of profits.
pip install systemr
Requires Python 3.9 or higher.
from systemr import SystemRClient
client = SystemRClient(api_key="sr_agent_...")
gate = client.pre_trade_gate(
symbol="AAPL",
direction="long",
entry_price="185.50",
stop_price="180.00",
equity="100000",
)
print(gate)
pre_trade_gate combines position sizing, risk validation, and supplied system-health context into a single decision-support response.
from systemr import SystemRClient
client = SystemRClient(api_key="sr_agent_...")
resp = client.chat(
"Review these R-multiples and tell me what changed in the system: 1.5, -1.0, 2.0, -0.5, 1.8"
)
print(resp["text"])
LLM-backed workflows may use credits depending on the live billing rules. Check the live pricing and billing surfaces before building production workflows.
Every current tool should be discovered from the live catalog before use:
tools = client.list_tools()
Generic tool calls:
result = client.call_tool(
"calculate_position_size",
equity="100000",
entry_price="185.50",
stop_price="180.00",
direction="long",
)
Common tool areas include:
The same API Toolkit can be used through MCP-compatible clients and REST integrations.
System R AI uses usage-based credits for paid workflows. Current rates and billing rules should be checked through the live pricing endpoint and the System R billing page.
Do not assume every tool, data path, or LLM-backed workflow has the same pricing behavior. Use live discovery and billing responses as the source of truth.
System R is software for decision support. Users remain responsible for their own trading and investing decisions. AI outputs can be wrong.
System R is not:
io.github.shelvick/shopify-subscription-reconciliation
zleventer/google-ads-mcp
csoai-org/meok-stripe-acp-checkout-mcp
io.github.mharnett/google-ads
csoai-org/stripe-billing-mcp
co.pipeboard/google-ads-mcp