# `software.brainiac/llms-txt` 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/software.brainiac/llms-txt
- JSON: https://radar.anlyon.com/api/v1/servers/software.brainiac/llms-txt
- Version: 1.0.1 (active)
- Publisher: software.brainiac (domain proven to the MCP Registry)
- Runs: remote (streamable-http)
- Sign-in to list tools: none
- Metadata from the official MCP Registry, synced 2026-11-20T01:18:20.821Z

Title and description, as the publisher wrote them:

```untrusted-server-text
BRAINIAC llms.txt Intelligence
Paid llms.txt Intelligence API via x402. Unpaid tool calls return 402.
```

Scanned 2026-11-20T01:19:14.654Z over streamable-http.

**8 tools: 2 change things and 6 only read.**

| Group | Class | Tools | Claude Code |
| --- | --- | --- | --- |
| Deletes or cancels | destructive | 0 | deny |
| Can't be undone | irreversible | 0 | ask |
| Changes things | write | 2 | ask |
| Only reads | read | 6 | allow |

## Exfil surface

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

| Tag | Meaning | Tools |
| --- | --- | --- |
| private-data | Reads files, mail, messages, databases, secrets, calendars or repositories. | `mcp__llms-txt__llms-txt_contradiction-detect` |
| untrusted-input | Brings in text someone else can write: web pages, inbound email, messages, issues and comments. | `mcp__llms-txt__llms-txt_context-budget-analyze` `mcp__llms-txt__llms-txt_contradiction-detect` `mcp__llms-txt__llms-txt_coverage-analyze` `mcp__llms-txt__llms-txt_instruction-risk-scan` `mcp__llms-txt__llms-txt_link-quality-check` `mcp__llms-txt__llms-txt_navigation-analyze` `mcp__llms-txt__llms-txt_structure-validate` `mcp__llms-txt__llms-txt_version-consistency` |
| outbound | Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or URL. | none |

## Findings

- **warn** Annotations that disagree with the tool in `llms-txt_link-quality-check` ([annotation-conflict](https://radar.anlyon.com/methodology#annotation-conflict))

```untrusted-server-text
The server marks "llms-txt_link-quality-check" read-only (readOnlyHint: true), and its own name, description or input give the class "write".

annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":true}; The name "llms-txt_link-quality-check" says "link": a word for a change.
```


## Changes things (2)

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

### `llms-txt_instruction-risk-scan`

- 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
Look in caller-supplied llms.txt text for a fixed set of instruction-override phrases, script tags, hidden characters, and javascript links. Ordinary how-to steps are not flagged, and nothing is executed. Price $0.04 USDC. POST /api/v1/llms-txt/instruction-risk-scan. An unpaid call returns the x402 payment terms and does not run.
```

### `llms-txt_link-quality-check`

- The name "llms-txt_link-quality-check" says "link": a word for a change.
- 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.
- Findings: warn annotation-conflict

Description, as the server wrote it:

```untrusted-server-text
Review link titles and descriptions in caller-supplied llms.txt content for generic labels and missing descriptions. Links are not opened. Price $0.03 USDC. POST /api/v1/llms-txt/link-quality-check. An unpaid call returns the x402 payment terms and does not run.
```

## Only reads (6)

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

### `llms-txt_context-budget-analyze`

- The name "llms-txt_context-budget-analyze" says "analyze": a word for reading.
- 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
Measure caller-supplied llms.txt size, repeated lines, and repeated link paths. The token figure is characters divided by four, not a model tokenizer. Price $0.03 USDC. POST /api/v1/llms-txt/context-budget-analyze. An unpaid call returns the x402 payment terms and does not run.
```

### `llms-txt_contradiction-detect`

- The description's first verb is "compare": a word for reading.
- Tag private-data: The description's first sentence says "compare" and "document": it reads documents.
- Tag untrusted-input: The server marks it openWorldHint: true: it reaches things outside the server's own data. The description's first sentence says "compare" and "http": it reads web pages, which anyone can write.

Description, as the server wrote it:

```untrusted-server-text
Compare caller-supplied llms.txt text with caller-supplied documents for conflicting account, API key, price, and HTTP method statements. This is a phrase match, not a reading of intent. Price $0.04 USDC. POST /api/v1/llms-txt/contradiction-detect. An unpaid call returns the x402 payment terms and does not run.
```

### `llms-txt_coverage-analyze`

- The name "llms-txt_coverage-analyze" says "analyze": a word for reading.
- 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
Compare llms.txt link paths with a caller-supplied page list and report pages that have no matching link. Pages are not fetched, and an unmatched link is not treated as a missing site page. Price $0.04 USDC. POST /api/v1/llms-txt/coverage-analyze. An unpaid call returns the x402 payment terms and does not run.
```

### `llms-txt_navigation-analyze`

- The name "llms-txt_navigation-analyze" says "analyze": a word for reading.
- 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
Find repeated link titles that point at different paths and generic section names in caller-supplied llms.txt content. No page is fetched. Price $0.03 USDC. POST /api/v1/llms-txt/navigation-analyze. An unpaid call returns the x402 payment terms and does not run.
```

### `llms-txt_structure-validate`

- The name "llms-txt_structure-validate" says "validate": a word for reading.
- 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
Check caller-supplied llms.txt text or structured JSON for one H1, a summary blockquote, H2 sections, and well-formed list links. URLs are not fetched. Price $0.03 USDC. POST /api/v1/llms-txt/structure-validate. An unpaid call returns the x402 payment terms and does not run.
```

### `llms-txt_version-consistency`

- The description's first verb is "compare": a word for reading.
- 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
Compare caller-supplied llms.txt versions for /api/vN/ majors, summary semantic versions, updated dates, and link titles whose paths differ. Neither version is fetched. Price $0.04 USDC. POST /api/v1/llms-txt/version-consistency. An unpaid call returns the x402 payment terms and does not run.
```

## Claude Code permissions

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

```json
{
  "permissions": {
    "allow": [
      "mcp__llms-txt__llms-txt_structure-validate",
      "mcp__llms-txt__llms-txt_context-budget-analyze",
      "mcp__llms-txt__llms-txt_coverage-analyze",
      "mcp__llms-txt__llms-txt_navigation-analyze",
      "mcp__llms-txt__llms-txt_contradiction-detect",
      "mcp__llms-txt__llms-txt_version-consistency"
    ],
    "ask": [
      "mcp__llms-txt__llms-txt_link-quality-check",
      "mcp__llms-txt__llms-txt_instruction-risk-scan"
    ],
    "deny": []
  }
}
```

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

- `mcp__llms-txt__llms-txt_structure-validate`: untrusted-input
- `mcp__llms-txt__llms-txt_link-quality-check`: untrusted-input
- `mcp__llms-txt__llms-txt_context-budget-analyze`: untrusted-input
- `mcp__llms-txt__llms-txt_coverage-analyze`: untrusted-input
- `mcp__llms-txt__llms-txt_navigation-analyze`: untrusted-input
- `mcp__llms-txt__llms-txt_contradiction-detect`: private-data, untrusted-input
- `mcp__llms-txt__llms-txt_instruction-risk-scan`: untrusted-input
- `mcp__llms-txt__llms-txt_version-consistency`: untrusted-input

## Remote posture

What an anonymous client saw at the remote endpoint, last 2026-11-20T19:33:51.516Z, unchanged since 2026-11-20T19:33:51.516Z. 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 | `2` |
| 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 2 tools that change things (2 write). Whether a call also works without credentials was not tested: tools are never called.

https://www.brainiac.software/mcp/llms-txt: tools/list returned 8 tools to an anonymous client, including llms-txt_link-quality-check (write), llms-txt_instruction-risk-scan (write)
```
