# No look-alike letters from other scripts

**A homoglyph uses a letter from another script that looks identical (Cyrillic а for Latin a) so a tool name or word imitates another; tool names should be plain ASCII and every other word should use one script.**

Check `sec-homoglyph` · Poisoning and hidden text · default severity: blocking (or attention, depending on the case) · rubric 2.1.0. Last verified against OpenAI plugin guidelines, section 3.3 on 2026-11-18.

## What fails

A tool name with non-ASCII letters (for example a Cyrillic "а" in "reаd_file") can impersonate another tool. In a name this is blocking; a word that mixes Latin with Cyrillic or Greek letters in a title or description is marked for attention.

## The rule

> Plugin and tool metadata must not seek secrets or protected context, direct unauthorized data transfers or actions, override platform instructions or safeguards, conceal behavior, or impersonate another party.

Source: [OpenAI plugin guidelines, section 3.3](https://developers.openai.com/plugins/plugin-guidelines)

## How to fix it

Use plain ASCII letters in names, and one script per word elsewhere.

1. Replace non-ASCII letters in tool names with ASCII ones.
2. In titles and descriptions, keep each word in one script.
3. Add a CI test that fails on non-ASCII tool names.

A CI test that fails when the pattern comes back:

### TypeScript (node:test)

```js
// tools-list.test.mjs: run `node --test` after saving the answer with
// npx @modelcontextprotocol/inspector --cli <server> --method tools/list > tools.json
import assert from 'node:assert/strict';
import { readFileSync } from 'node:fs';
import { test } from 'node:test';

const answer = JSON.parse(readFileSync('tools.json', 'utf8'));
const tools = answer.tools ?? answer.result?.tools ?? answer;

// Every string and every key in a tool, with its path.
function* strings(value, path) {
  if (typeof value === 'string') yield [path, value];
  else if (value && typeof value === 'object')
    for (const [k, v] of Object.entries(value)) {
      yield [`${path}.${k} (key)`, k];
      yield* strings(v, `${path}.${k}`);
    }
}

const PATTERN = /[^\x00-\x7F]/u;

test('tool names are plain ASCII', () => {
  const hits = [];
  for (const tool of tools) if (PATTERN.test(tool.name)) hits.push(tool.name);
  assert.deepEqual(hits, [], 'tool names with non-ASCII letters can impersonate other tools');
});
```

### Python (pytest)

```python
# test_tools_list.py: run `pytest` after saving the answer with
# npx @modelcontextprotocol/inspector --cli <server> --method tools/list > tools.json
import json
import re

with open("tools.json", encoding="utf-8") as f:
    answer = json.load(f)
tools = answer.get("tools") or answer.get("result", {}).get("tools") if isinstance(answer, dict) else answer


def strings(value, path):
    """Every string and every key in a tool, with its path."""
    if isinstance(value, str):
        yield path, value
    elif isinstance(value, dict):
        for k, v in value.items():
            yield f"{path}.{k} (key)", k
            yield from strings(v, f"{path}.{k}")
    elif isinstance(value, list):
        for i, v in enumerate(value):
            yield from strings(v, f"{path}.{i}")


PATTERN = re.compile(r"[^\x00-\x7F]")


def test_tool_names_are_plain_ascii():
    hits = [tool["name"] for tool in tools if PATTERN.search(tool["name"])]
    assert hits == [], "tool names with non-ASCII letters can impersonate other tools"
```

## Questions

### How can a tool name impersonate another tool?

Two names that render identically are different strings. A malicious server can list reаd_file (with a Cyrillic а) next to a trusted read_file; a person approving the call sees the trusted name.

### Are non-English descriptions flagged?

No. The check flags words that mix Latin with Cyrillic or Greek letters, which ordinary text in any one language does not do.

## Related checks

- [Allowed characters in MCP tool names](https://isokron.ai/mcp-check/checks/spec-name-charset)
- [No invisible Unicode tag or variation-selector characters](https://isokron.ai/mcp-check/checks/sec-invisible-encoding)
- [No steering of when other tools are used](https://isokron.ai/mcp-check/checks/sec-cross-tool-steering)
- [Tool poisoning and hidden text](https://isokron.ai/mcp-check/checks/security)

Counts toward: Poisoning and hidden text, Anthropic connector directory, OpenAI plugin directory.

Rubric: https://isokron.ai/mcp-check/rubric

Check your tool list: https://isokron.ai/mcp-check
