Markets and digital assets

Trading and capital markets

Agents that trade must stay inside limits that the firm, not the model, sets.

Firms use AI for research, order routing and trading, and agents can now act across these steps. IOSCO's consultation report on AI in capital markets points to risks such as herding behaviour, model and data issues, and dependence on a small number of third-party providers. AgenticOpenFinance™ puts position limits, instrument lists and kill switches in front of every agent order and keeps the record.

What agents do

What you build on AgenticOpenFinance™

1

Trading mandates

Instruments, venues, order sizes and daily exposure are set per agent and checked before each order.

2

Pre-trade policy

Rules check restricted lists, concentration and market conditions and return allow, deny or escalate.

3

Kill switch

A desk head can halt one agent or all agents with one action.

4

Model change control

A new model version runs only after validation and is recorded with each order.

5

Order evidence

Every order keeps the inputs, the policy version, the verdict and the execution report.

Value chain

The actors in this chain

AgenticOpenFinance™ connects them with one identity check, one mandate model, one policy and one evidence trail.

Portfolio managerTrading agentRisk managerBrokerVenueCompliance

Regulators and standards in the directory

Typical flow

A governed agent flow, step by step

  1. Signal
  2. Order drafted
  3. Mandate and limits
  4. Pre-trade policy
  5. Escalation if needed
  6. Execution
  7. Evidence and surveillance

Live demo

See it in the terminal

A multi-agent invoice workflow with a policy verdict and the finance lead's approval (demo). Demo data

DemoAll companies, people, agents and amounts are fictional. Protocol names refer to public specifications; naming one is not a partnership or an endorsement.

Glossary

Key terms for this sector

Build Trading and capital markets on AgenticOpenFinance™.