TRUE Documentation

What is TRUE

TRUE is an AI-native financial SuperApp. It pairs a multi-agent reasoning loop with a multi-source price engine and a survivability-aware execution layer, surfaced through a streaming chat experience and exposed end-to-end through MCP. The same tools the TRUE app uses internally are the tools you can call from outside.

These docs are the single source of truth for how every TRUE feature works — for users who want to use the product safely, for partners distributing TRUE inside their own surfaces, and for developers building on top.

Architecture at a glance

The TRUE platform is built around three composable layers. Each layer is documented as its own section and is replaceable in isolation: a better LLM, a better oracle, or a better execution venue can be swapped in without touching the others.

Quickstart

Start at the section that matches what you’re building. Every doc is self-contained — you don’t need to read them in order.

• Building on top of TRUE? Start with MCP — every tool TRUE uses internally is exposed there.
• Integrating prices? See Price Engine for source priority and failover.
• Curious how the agents work? Agents walks through the DARS pipeline.
• Routing trades through us? Agentic Trading covers SAE and the 5-tier liquidity stack.
• New to the product? Read Getting Started followed by Safety Overview.

How to read these docs

Every page answers four questions in order: what is this, how does it actually work, how do I use it safely, and how do I integrate, configure, or extend it. Where relevant, audiences are split into For Users, For Partners, and For Developers so you can jump to your slice.

Code blocks use dual themes that follow your light/dark preference. Anchor links are stable. Every page ships as static HTML — no client routing, no hydration cost.

Safety first

Before you do anything else, read the Safety Overview. TRUE never holds keys, never DMs first, and never executes without an explicit user signature. Knowing the boundaries up front saves you the cost of learning them the hard way.