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

# Quickstart (MCP)

### Wire it up in one minute

RAILHOOD speaks MCP — the open standard AI agents use to discover and call tools. Tools are free to call; the protocol takes its cut inside the transaction, like any venue. No accounts. No API keys.

### Connect

#### Remote

```
claude mcp add --transport http railhood https://railhood.com/mcp
```

#### Local (stdio)

```
claude mcp add --transport stdio railhood -- npx -y railhood-privacy
```

No MCP client? Same actions from the command line. If your agent has shell access, it has RAILHOOD.

### Current release

| Spec           | v1                                                                     |
| -------------- | ---------------------------------------------------------------------- |
| Asset          | ETH                                                                    |
| Standard sizes | 0.001 · 0.01 · 0.1 · 1 · 10 · 100 ETH                                  |
| Protocol fee   | 0.25% in · 0.25% out                                                   |
| Exit gas       | Paid by the settlement agent — your wallet never appears as the sender |
| Custody        | None. You sign your own transactions.                                  |

### The tools

| API verb     | What it does                                                                                                                                                                            |
| ------------ | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| `pool_stats` | Reads the crowd: pool depth per standard size, live. Bigger crowd, stronger privacy — check it before you commit. How to read it: [Privacy quality](/railhood-docs/privacy-quality.md). |
| `shield`     | Takes a position off-market — deposits ETH into the pool as a sealed note, in a standard size.                                                                                          |
| `unshield`   | Settles back on-market — exits to any address through the gasless settlement agent.                                                                                                     |

The API verbs keep the crypto-native names; the docs use market language because that's what the actions are. More tools land per the [Roadmap](/railhood-docs/roadmap.md).

### Your first pass, end to end

* **1 — Read the crowd.** Call `pool_stats`. You want depth in your size: a 1 ETH position in a deep 1-ETH pool is a face in a crowd; in a shallow one it's a face in a hallway.
* **2 — Go off-market.** Call `shield` with a standard size. Your funds become a sealed note; the tape records only that *someone* entered at that size.
* **3 — Sit.** Time is camouflage. Entering and exiting in the same breath telegraphs the link — treat the pool like a position, not a corridor. More in [Privacy quality](/railhood-docs/privacy-quality.md).
* **4 — Settle out.** Call `unshield` to any address — a fresh one, an exchange deposit, a counterparty. The settlement agent submits and pays gas; the tape never prints your original wallet.
* **5 — Back up the note.** It's a bearer instrument. Lose it, lose the funds. Treat it exactly like a private key, because it is one.

### Why standard sizes

Deposits come in a fixed menu on purpose. If everyone entered random amounts, an observer could match a 12.4073-ETH entry to a 12.4073-ETH exit and the privacy is toast. Standard sizes make every entry look identical. TradFi solved the same problem the same way a century ago — it's why shares trade in round lots and cash comes in denominations. Uniformity is what makes a crowd a crowd.


---

# 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 dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://railhood.gitbook.io/railhood-docs/quickstart-mcp.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

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.