Skip to content
IsoKron

No HTML comments in metadata

An HTML comment (<!-- ... -->) is hidden wherever a description is rendered as HTML or Markdown, but the model reads it, so it can carry instructions no reviewer sees; both directories reject hidden instructions.

Check sec-html-comment · Poisoning and hidden text · default severity: blocking · rubric 2.1.0

What fails

An HTML comment is hidden wherever the description is rendered as HTML or Markdown, but the model reads it.

The rule

Contain hidden, obfuscated, or encoded instructions

Source: Anthropic connector pre-submission checklist

How to fix it

Remove the comment. Anything the model should know belongs in visible text.

  1. Remove every <!-- ... --> from names, titles, descriptions and schemas.
  2. Put anything the model should know in visible text.
  3. Add a CI test.

A CI test that fails when the pattern comes back, run against the saved tools/list answer:

TypeScript (node:test)

TypeScript test
// 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 = /<!--/;

test('no HTML comments', () => {
  const hits = [];
  for (const tool of tools) for (const [path, text] of strings(tool, tool.name)) if (PATTERN.test(text)) hits.push(path);
  assert.deepEqual(hits, [], 'HTML comments hide text from reviewers');
});

Python (pytest)

Python test
# 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"<!--")


def test_no_html_comments():
    hits = [path for tool in tools for path, text in strings(tool, tool["name"]) if PATTERN.search(text)]
    assert hits == [], "HTML comments hide text from reviewers"

Questions

Our descriptions are generated from Markdown docs with comments. Is that a problem?

Yes for the tool list: strip comments when you generate the descriptions. The model receives the raw text either way.

Is any HTML allowed?

Plain text is safest. Tags that look like prompt markup (<IMPORTANT>, <system>) are a separate blocking check.

How this check is graded

It counts toward: Poisoning and hidden text, Anthropic connector directory, OpenAI plugin directory. See the rubric.

Last verified against Anthropic connector pre-submission checklist on 2026-11-18. The rule above is quoted in the source’s own words. This page as Markdown.

Check your tool list: paste your tools/list, nothing is sent anywhere