We map agentic finance.
Tests of the tools builders actually use, case studies of what we build on the foundation, and analysis of where the ecosystem is heading. Nobody who is actually building on finance is writing this down, so we do.
Three tracks.
Real stories
What we build on the foundation, told plainly. The proof, not the pitch.
Teardowns
We take the tools builders actually use, connect them, and document exactly where they work and where they break.
The Map
Where agentic finance is heading, and what the still-forming regulation means for people building now.
Live now, and what is next.
A safety check for the trading tool you built with AI
Building a trading or investing tool with a coding agent takes an evening, and checking what that tool can do with your account takes about one more. These are the nine questions I'd go through, starting with the key and ending with the off switch.
InteractiveSpending limits for AI agents: soft, hard and signed
Card networks, payment platforms and fintechs have all announced ways to let an AI agent pay. They all set a budget, but they put it in very different places, and that place decides whether the limit holds.
InteractiveWhich brokers let AI trade your account, and how far each one goes
In 2026, broker after broker opened a door for AI assistants. Some let the AI look, some let it trade after you say yes, a few let it trade on its own. What matters most is where that yes is enforced.
InteractiveMy trading journal updates itself every night. Here is how I built it.
I asked an AI coding agent to build a private trading journal and performance dashboard from my MT5 history. It reads my trades, never touches an order, and rebuilds itself while I sleep.
InteractiveHow often should you rebalance? What the research says, and where AI helps
The research favours a rebalancing rule that needs daily attention; an AI assistant can supply the attention, as long as it never does the maths and never places the order.
InteractivePrompting Against the Machine: Using AI in Trading
How to prompt an AI about your own trading, whether you trade by hand, by rules, or by code it wrote for you.
InteractiveThe agent was ready. The bank wasn't.
I asked an AI agent to pay my kids' kindergarten invoices. It turned out to be the most careful party in the payment. The banking infrastructure around it was the weak part.
InteractiveSeven orders, six refusals
I spent an afternoon trying to make Vibe-Trading misbehave. Seven orders went in, six came back refused. A teardown of the safety layer, and a twenty-minute experiment you can run yourself.
InteractiveHand the agent the company account
My bank could not let an agent pay for me without my tap. A business account in Singapore can. Here is what that lets a small company do, and the one place a person still says yes.
Interactive
Two people, one finished foundation: how Takumis went from an evening idea to live trades
Takumis is a platform for vibe-building trading strategies. You talk your idea through with an AI, it tests it on data, and then you trade for real.
One piece at a time
Each report is a thing we actually built, tested or worked out, written up once it is worth reading.
Written by the people building it
The people who build it, with traders and engineers who work on this every day. Roughly one considered piece at a time.
New reports, slowly.
We publish when we have something real to show. Leave your email and we’ll send each new report when it lands.
Prefer email? hello@hello-purple.com