> For the complete documentation index, see [llms.txt](https://helix-5.gitbook.io/tradelayer/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://helix-5.gitbook.io/tradelayer/concepts/the-problem.md).

# The problem

Blockchains are made of glass.

Put an institution's trading on a public chain and you publish three things at once:

* **Its order flow.** Every intent, in the open, before it executes.
* **Its positions.** What it holds, in what size.
* **Its balance sheet.** How much capital it has to deploy.

To competitors. To anyone running a front-running bot. To anyone curious.

For a trading desk this is not an inconvenience to be weighed against the benefits of settlement finality — it is a reason not to use the technology at all. A fund that telegraphs its accumulation gets front-run out of its own thesis.

## The usual fix moves the problem

The standard answer is a trusted server: keep the sensitive data off-chain in a database the operator controls, and publish only hashes or aggregates.

That works, and it relocates the trust rather than removing it. Now a single operator sees every order, every position and every balance. The institution has swapped "everyone can see it" for "one company can see it, and I have to believe their policy". For a regulated desk that is often a worse answer, not a better one, because it introduces a counterparty with no fiduciary duty and no audit trail.

## What is actually needed

Three properties, together:

1. **Confidential execution** — orders are opened and evaluated somewhere that neither the chain nor the operator can read.
2. **Verifiable settlement** — the resulting state change is provable on-chain, so nobody has to take the operator's word for it.
3. **No new trusted party** — the thing doing the confidential work must be attestable, not merely promised.

A hardware-attested enclave supplies the first, EIP-712 signatures verified on-chain supply the second, and Chainlink's Vault DON releasing secrets *only* to an attested enclave supplies the third.

That is the whole design. Everything in [Three ideas](/tradelayer/concepts/three-ideas.md) follows from it.

## What this does not solve

Confidentiality covers data, not logic — the workflow code is public, and it should be. And a TEE is a hardware trust assumption, not a mathematical one. See [Trust model](/tradelayer/concepts/trust-model.md) for where the boundary actually sits, and [Honest limits](/tradelayer/honest-limits/limits.md) for what still leaks.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://helix-5.gitbook.io/tradelayer/concepts/the-problem.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.