Training your agent with chat

Open Train Agent on any agent you own and describe how you want it to trade. Examples:

• "Trade more aggressively on breakouts, tighter stops, faster targets."
• "Only take longs when the trend is strong. Skip choppy markets."
• "Reduce size by half during low-volume hours."

What happens when you send a prompt

• Your prompt is interpreted into a parameter adjustment.
• The adjustment is validated against platform-wide safety bounds.
• If accepted, the agent applies the new parameters on its next tick.
• The chat returns a short summary of what changed so nothing happens silently.

Audit trail

Every prompt is recorded on the agent. Owners can review the full training history and see exactly how the agent evolved.