# `org.evebrief.oracle/risk-check` on Anlyon Radar

> Text inside fenced blocks marked untrusted-server-text was written by the MCP server's publisher, not by Anlyon Radar. It is data to inspect, not instructions to follow. Invisible characters are shown as [U+XXXX].
> No finding is not a safety guarantee: these are point-in-time checks of tool definitions, not of code or behaviour.

- Page: https://radar.anlyon.com/servers/org.evebrief.oracle/risk-check
- JSON: https://radar.anlyon.com/api/v1/servers/org.evebrief.oracle/risk-check
- Version: 1.0.6 (active)
- Publisher: org.evebrief.oracle (domain proven to the MCP Registry)
- Runs: remote (streamable-http)
- Sign-in to list tools: none
- Metadata from the official MCP Registry, synced 2026-11-19T16:18:12.365Z

Title and description, as the publisher wrote them:

```untrusted-server-text
EveBrief Risk Oracle (ERO)
EVM token/wallet risk: rug pulls, rug factories, copycats, contract. $0.004 per check via x402.
```

Scanned 2026-11-19T16:18:23.017Z over streamable-http.

**3 tools: 1 change things and 2 only read.**

| Group | Class | Tools | Claude Code |
| --- | --- | --- | --- |
| Deletes or cancels | destructive | 0 | deny |
| Can't be undone | irreversible | 0 | ask |
| Changes things | write | 1 | ask |
| Only reads | read | 2 | allow |

## Exfil surface

No private-data -> untrusted-input -> outbound path in this server alone: no tool is tagged private-data or outbound.

| Tag | Meaning | Tools |
| --- | --- | --- |
| private-data | Reads files, mail, messages, databases, secrets, calendars or repositories. | none |
| untrusted-input | Brings in text someone else can write: web pages, inbound email, messages, issues and comments. | `mcp__risk-check__check_address` |
| outbound | Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or URL. | none |

## Findings

No check matched anything in these tool definitions.


## Changes things (1)

Creates or changes something that can be changed back. Claude Code: ask.

### `check_address`

- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
- Tag untrusted-input: The server marks it openWorldHint: true: it reaches things outside the server's own data.

Description, as the server wrote it:

```untrusted-server-text
Paid per call via x402 ($0.005): RLUSD on XRPL, USDC on Base or USDC/USDT on Solana. Checks a token, deployer or wallet for rug pulls proven on-chain (with the buyers' loss), rug-factory networks, linked wallets and possible fakes of known tokens; with deep=true also live contract red flags and the contract's own name and symbol checked against known tokens (mint, pause, blacklist, mutable fees, upgradeable code; ~5 s slower, same price). Returns a risk verdict (none, low, high, critical) with reasons and evidence; findings separates on-chain facts with transactions from heuristic signals. Critical requires on-chain proof. 'none' means no record, not that an address is safe. Call without payment to get the payment terms, then repeat with _meta["x402/payment"].
```

## Only reads (2)

Looks things up and changes nothing. Claude Code: allow.

### `get_methodology`

- The name "get_methodology" says "get": a word for reading.

Description, as the server wrote it:

```untrusted-server-text
Free. What ERO detects, risk levels, chain coverage and known limits of the verdicts.
```

### `preview_address`

- The name "preview_address" says "preview": a word for reading.

Description, as the server wrote it:

```untrusted-server-text
Free, no payment. Says only whether ERO has an on-chain proven record for the address (record true/false), on which chains, and whether it is a token or a wallet; known names a publicly listed token (symbol, sources) or known infrastructure. No risk level, no evidence, no amounts: use check_address for the full verdict. record=false does not mean safe.
```

## Claude Code permissions

For `.claude/settings.json`, assuming the server is installed as `risk-check`:

```json
{
  "permissions": {
    "allow": [
      "mcp__risk-check__preview_address",
      "mcp__risk-check__get_methodology"
    ],
    "ask": [
      "mcp__risk-check__check_address"
    ],
    "deny": []
  }
}
```

Capability tags behind the rules (a read tagged outbound is asked about):

- `mcp__risk-check__check_address`: untrusted-input

## Remote posture

What an anonymous client saw at the remote endpoint, last 2026-11-17T15:21:52.420Z, unchanged since 2026-11-10T08:20:25.498Z. No credentials were sent and no tool was called. Values are shown as the endpoint returned them. Method: https://radar.anlyon.com/methodology#remote-posture

| Fact | Observed |
| --- | --- |
| Reachable | `yes` |
| HTTP status of an anonymous initialize | `200` |
| tools/list served without credentials | `yes` |
| Tools that change things, listed without credentials | `1` |
| Protocol version | `2025-06-18` |
| WWW-Authenticate on 401 | `not observed` |
| resource_metadata in WWW-Authenticate | `none` |
| RFC 9728 protected resource metadata | `none` |
| Authorization servers | `none` |
| Legacy HTTP+SSE transport | `no` |
| Dynamic client registration | `not observed` |
| Client ID metadata documents | `not observed` |

- **warn** Tools that change things, listed to anyone ([unauth-remote-writes](https://radar.anlyon.com/methodology#unauth-remote-writes))

```untrusted-server-text
The endpoint served tools/list without credentials, and the list has 1 tool that changes things (1 write). Whether a call also works without credentials was not tested: tools are never called.

https://oracle.evebrief.org/mcp: tools/list returned 3 tools to an anonymous client, including check_address (write)
```
