System R AI

1. Install

pip install systemr

Requires Python 3.9 or higher.

2. Register

curl -X POST https://agents.systemr.ai/v1/agents/register \
  -H "Content-Type: application/json" \
  -d '{
    "owner_id": "me",
    "agent_name": "MyBot",
    "agent_type": "trading"
  }'

Response:

{
  "agent_id": "agt_a1b2c3d4",
  "agent_name": "MyBot",
  "agent_type": "trading",
  "mode": "sandbox",
  "api_key": "sr_agent_abc123def456...",
  "created_at": "2026-03-25T12:00:00Z"
}

Save the api_key. It is shown once and cannot be retrieved later.

3. Call

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(f"Gate passed: {gate['gate_passed']}")
print(f"Shares: {gate['sizing']['shares']}")
print(f"Risk amount: ${gate['sizing']['risk_amount']}")

This single call runs position sizing (G-formula), risk validation (Iron Fist rules), and system health assessment. Cost: $0.01.

What just happened

The SDK sent POST /v1/compound/pre-trade-gate with your parameters.
System R sized the position using the G-formula (geometric growth optimization).
Iron Fist risk rules validated the trade against position risk, portfolio risk, and daily loss limits.
$0.01 was deducted from your compute credit balance.
You received a complete pre-trade report in one response.

Next steps

Browse all 55 tools available to your agent
Connect via MCP for Claude Desktop, Cursor, or any MCP client
View pricing for per-call costs
Connect a broker to execute trades