# `ai.coreyjames/portfolio` 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/ai.coreyjames/portfolio
- JSON: https://radar.anlyon.com/api/v1/servers/ai.coreyjames/portfolio
- Version: 1.0.0 (active)
- Publisher: ai.coreyjames (domain proven to the MCP Registry)
- Runs: remote (streamable-http)
- Sign-in to list tools: none
- Metadata from the official MCP Registry, synced 2026-10-28T00:17:38.698Z

Title and description, as the publisher wrote them:

```untrusted-server-text
Corey James portfolio
Fit reports and search over Corey James's work: GPU inference, agent governance, enterprise AI.
```

Scanned 2026-11-09T20:18:06.192Z over streamable-http.

**13 tools: 1 do something that can't be undone, 3 change things and 9 only read.**

| Group | Class | Tools | Claude Code |
| --- | --- | --- | --- |
| Deletes or cancels | destructive | 0 | deny |
| Can't be undone | irreversible | 1 | ask |
| Changes things | write | 3 | ask |
| Only reads | read | 9 | allow |

## Exfil surface

6 private-data -> untrusted-input -> outbound paths in this server alone. Denying 1 private-data tool breaks every one: `mcp__portfolio__fetch`.

| Tag | Meaning | Tools |
| --- | --- | --- |
| private-data | Reads files, mail, messages, databases, secrets, calendars or repositories. | `mcp__portfolio__fetch` |
| untrusted-input | Brings in text someone else can write: web pages, inbound email, messages, issues and comments. | `mcp__portfolio__fetch` `mcp__portfolio__get_page` `mcp__portfolio__search` |
| outbound | Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or URL. | `mcp__portfolio__contact` `mcp__portfolio__get_page` |

| Private data | Untrusted input | Outbound |
| --- | --- | --- |
| `mcp__portfolio__fetch` | `mcp__portfolio__fetch` | `mcp__portfolio__contact` |
| `mcp__portfolio__fetch` | `mcp__portfolio__fetch` | `mcp__portfolio__get_page` |
| `mcp__portfolio__fetch` | `mcp__portfolio__get_page` | `mcp__portfolio__contact` |
| `mcp__portfolio__fetch` | `mcp__portfolio__get_page` | `mcp__portfolio__get_page` |
| `mcp__portfolio__fetch` | `mcp__portfolio__search` | `mcp__portfolio__contact` |
| `mcp__portfolio__fetch` | `mcp__portfolio__search` | `mcp__portfolio__get_page` |

## Findings

No check matched anything in these tool definitions.


## Can't be undone (1)

Sends, pays, publishes or merges: once done, it cannot be taken back. Claude Code: ask.

### `contact`

- The description's first verb is "send": a word for sending, and what is sent cannot be taken back.
- Tag outbound: The description's first verb is "send": it sends or publishes something to other people.

Description, as the server wrote it:

```untrusted-server-text
Send a message to Corey James through the site's contact form, on behalf of the user. Requires the user's own name and email.
```

## Changes things (3)

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

### `match_role`

- Nothing in its name or description says what it does, so it is counted as a write.

Description, as the server wrote it:

```untrusted-server-text
Given a job description or a list of requirements, return an evidence-backed fit report: for each requirement, the pages and figures on coreyjames.ai that demonstrate it, plus the gaps. Cites, never asserts.
```

### `scale`

- Nothing in its name or description says what it does, so it is counted as a write.

Description, as the server wrote it:

```untrusted-server-text
The size of the body of work: source lines, commits, languages, systems, stories, ideas, tools, patent applications.
```

### `who`

- Nothing in its name or description says what it does, so it is counted as a write.

Description, as the server wrote it:

```untrusted-server-text
Who this is: name, the name used on patent filings, role, links and how to make contact.
```

## Only reads (9)

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

### `fetch`

- The name "fetch" says "fetch": a word for reading.
- Tag private-data: The description's first sentence says "fetch" and "document": it reads documents.
- Tag untrusted-input: The description's first sentence says "fetch" and "url": it reads web pages, which anyone can write.

Description, as the server wrote it:

```untrusted-server-text
Fetch one portfolio document by id (the page path from search): id, title, text (the page as Markdown), url, metadata.
```

### `get_page`

- The name "get_page" says "get": a word for reading.
- Tag untrusted-input: The description's first sentence says "url": it reads web pages, which anyone can write. The input takes a URL (input.url), and the tool reads it: it reads web pages, which anyone can write.
- Tag outbound: The input takes a URL (input.url): a request to a URL can carry data to whoever runs it.
- Claude Code: ask, because it is tagged outbound.

Description, as the server wrote it:

```untrusted-server-text
The record for one page by URL: title, result, systems, technologies, stats, lede, quote, evidence.
```

### `list_capabilities`

- The name "list_capabilities" says "list": a word for reading.

Description, as the server wrote it:

```untrusted-server-text
The capabilities the work demonstrates.
```

### `list_ideas`

- The name "list_ideas" says "list": a word for reading.

Description, as the server wrote it:

```untrusted-server-text
The recurring engineering principles.
```

### `list_patent_applications`

- The name "list_patent_applications" says "list": a word for reading.

Description, as the server wrote it:

```untrusted-server-text
The provisional patent applications (patent pending), with application numbers.
```

### `list_systems`

- The name "list_systems" says "list": a word for reading.

Description, as the server wrote it:

```untrusted-server-text
The ten systems with one-line results.
```

### `list_technologies`

- The name "list_technologies" says "list": a word for reading.

Description, as the server wrote it:

```untrusted-server-text
Technologies, optionally by kind (language, discipline, platform, framework, protocol, runtime, datastore, hardware, model).
```

### `search`

- The name "search" says "search": a word for reading.
- Tag untrusted-input: The description's first sentence says "search" and "url": it reads web pages, which anyone can write.

Description, as the server wrote it:

```untrusted-server-text
Search the portfolio (connector-compatible): returns results with id, title and url. Use fetch with an id for the full document.
```

### `search_portfolio`

- The name "search_portfolio" says "search": a word for reading.

Description, as the server wrote it:

```untrusted-server-text
Search the systems, engineering stories, capabilities, ideas, technologies and patent applications. Returns ranked pages with their measured results and links.
```

## Claude Code permissions

For `.claude/settings.json`, assuming the server is installed as `portfolio`:

```json
{
  "permissions": {
    "allow": [
      "mcp__portfolio__search_portfolio",
      "mcp__portfolio__search",
      "mcp__portfolio__fetch",
      "mcp__portfolio__list_systems",
      "mcp__portfolio__list_capabilities",
      "mcp__portfolio__list_ideas",
      "mcp__portfolio__list_technologies",
      "mcp__portfolio__list_patent_applications"
    ],
    "ask": [
      "mcp__portfolio__match_role",
      "mcp__portfolio__get_page",
      "mcp__portfolio__who",
      "mcp__portfolio__scale",
      "mcp__portfolio__contact"
    ],
    "deny": []
  }
}
```

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

- `mcp__portfolio__get_page`: untrusted-input, outbound
- `mcp__portfolio__search`: untrusted-input
- `mcp__portfolio__fetch`: private-data, untrusted-input
- `mcp__portfolio__contact`: outbound

## Remote posture

What an anonymous client saw at the remote endpoint, last 2026-11-19T17:23:40.813Z, unchanged since 2026-10-28T20:18:17.750Z. 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 | `4` |
| Protocol version | `2025-06-18` |
| WWW-Authenticate on 401 | `not observed` |
| resource_metadata in WWW-Authenticate | `none` |
| RFC 9728 protected resource metadata | `https://coreyjames.ai/.well-known/oauth-protected-resource` |
| Authorization servers | `none` |
| Legacy HTTP+SSE transport | `no` |
| Dynamic client registration | `not observed` |
| Client ID metadata documents | `not observed` |

- **high** 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 4 tools that change things (1 irreversible, 3 write). Whether a call also works without credentials was not tested: tools are never called.

https://coreyjames.ai/mcp: tools/list returned 14 tools to an anonymous client, including contact (irreversible), match_role (write), who (write), scale (write)
```
