Trade Execution Audit Trail

Every trade your AI agent executes, rejects, or flags: logged, hashed, and independently verifiable. Nine flag subtypes including position limit breach, wash trade risk, and restricted security. Your order and counterparty data stays in your environment.

Request pilot access Methodology for auditors

AI agents are in the order workflow. Most have no audit trail.

Trading desks and asset managers deploy AI agents to run pre-trade checks and execute or hold orders within position, risk, and regulatory limits. When a regulator or internal audit asks how the firm governs these agents, most firms cannot reconstruct the decision after the fact.

MiFID II Article 25 and SEC Rule 17a-4 require records that survive that question. An append-only, cryptographically verifiable decision log gives auditors a tamper-evident record of agent behavior without exposing proprietary order flow.

What gets logged for every decision

The decision

Execute, reject, or flag. Recorded with timestamp, agent ID, instrument reference, side, quantity, and notional as submitted by the agent.

The input fingerprint

A SHA-256 hash of the pre-trade check inputs: order reference, instrument, side, quantity, and limit checked. Raw order book data and client identities stay in your environment.

The rationale

The agent's stated reason, recorded as an immutable string. "Order would breach desk position limit for AAPL. Pre-trade check failed on notional threshold."

Flag classification

When the agent flags, Sigmodx records the subtype (position limit breach, wash trade risk, restricted security, and six others) and severity. Regulatory escalations route for immediate review.

How the agent's state is determined

Human review
Supervisors review flagged and rejected orders. Assessments feed execution accuracy, rejection precision, false positive rate, regulatory flag rate, and human override rate.
State assignment
Human override rate carries the most weight. Above 8% moves the agent to LIMIT; above 15% to BLOCK. False positive rate above 12% also triggers LIMIT.
State changes are logged
Every state change and the reason for it is written to the append-only log. Supervisors can manually override state with a mandatory reason field.

What you give your auditors

Your verification string

SIGMODX-TRADE-E54ECE50-D331D86614C1

Your auditors submit this string to sigmodx.com/verify. They receive cryptographic confirmation that the audit record for the attested period is intact and unaltered.

They get proof. They don't get your data.

You retain all underlying order records internally and produce them directly to regulators if required. Sigmodx is the proof layer.

Integration is one afternoon

Install the SDK

pip install sigmodx

npm install @sigmodx/sdk

Add three lines to your agent

Direct SDKLangChainMCP

pip install sigmodx

from sigmodx import SigmodxClient

client = SigmodxClient(api_key="...", agent_id="...")

input_hash = client.hash_inputs({
    "order_ref": "ORD-2026-7781",
    "symbol": "AAPL",
    "side": "buy",
    "quantity": 5000,
})

result = client.submit_trade_decision(
    decision_type="flag",
    input_hash=input_hash,
    rationale="Order would breach the desk position limit for this name.",
    flag_subtype="position_limit_breach",
    severity="high",
    instrument_symbol="AAPL",
    side="buy",
    quantity=5000,
    regulatory_escalation=True,
)

pip install sigmodx sigmodx-integrations

from sigmodx import SigmodxClient
from sigmodx_integrations import SigmodxCallbackHandler, TRADE_EXECUTION_CONFIG

client = SigmodxClient(api_key="...", agent_id="...")
handler = SigmodxCallbackHandler(
    client=client,
    config=TRADE_EXECUTION_CONFIG,
)

# Every tool call is automatically logged to Sigmodx
agent_executor.invoke(
    {"input": "Pre-trade check on order ORD-2026-7781"},
    config={"callbacks": [handler]},
)

pip install sigmodx-mcp

SIGMODX_API_KEY=your-key \
SIGMODX_AGENT_ID=your-agent-uuid \
sigmodx-mcp

Claude Desktop / Cursor config:

{
  "mcpServers": {
    "sigmodx": {
      "command": "sigmodx-mcp",
      "env": {
        "SIGMODX_API_KEY": "your-api-key",
        "SIGMODX_AGENT_ID": "your-agent-uuid"
      }
    }
  }
}

Decision events appear in your org dashboard immediately
No infrastructure changes. No vendor access to your trading systems.

Agent API reference →

Available now

The trade execution scenario is live in production. Pilot access is available for Q3 2026. 90 days, no charge, no contract.

Request pilot access →See how verification works →