MCP server for China advertising-law compliance check: flag prohibited wording in marketing copy.
Maintain this server? Claim it
Call this server's tools from your browser. Nothing to install.
Requests go through mcptry because browsers can't call most MCP servers directly. Header values are sent to the server for that request only and are never stored or logged.
Tool list from our last check. Connect live to refresh it.
查看广告法违禁词库：返回全部词条或指定 category 类别下的词表（词、类别、风险级别）。可选类别：绝对化用语、医疗功效、权威背书、诱导抢购、封建迷信、投资承诺、教育培训、房产违规、虚假夸大。
可选类别名，省略返回全部类别样本
This tool isn't marked read-only, so it may change something on the server or in an account you connected.
Pick your client. Values in capitals, like YOUR_TOKEN, are yours to fill in.
Run in a terminal
claude mcp add --transport http adcheck-mcp https://pianam.cn/adcheck-mcp/mcp
查看广告法违禁词库：返回全部词条或指定 category 类别下的词表（词、类别、风险级别）。可选类别：绝对化用语、医疗功效、权威背书、诱导抢购、封建迷信、投资承诺、教育培训、房产违规、虚假夸大。
| Argument | Type | Description |
|---|---|---|
| category |
| any |
| 可选类别名，省略返回全部类别样本 |
{
"type": "object",
"properties": {
"category": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"default": null,
"description": "可选类别名，省略返回全部类别样本"
}
},
"additionalProperties": false
}广告法违禁词/极限词检测。输入广告文案 text（商品标题、详情页、宣传语等），纯本地词库匹配返回：总体风险结论（通过/需修改/不通过）、命中违禁词清单（词、类别、风险级别、风险说明、法律依据、修改建议、原文上下文）、高中低风险数量统计、类别统计，以及自动打码后的安全文案。覆盖绝对化用语（国家级/最佳/第一/顶级）、医疗功效（治疗/根治/减肥）、权威背书（特供/免检）、诱导抢购（秒杀/万人疯抢）、封建迷信、投资承诺（保本/零风险）、教育培训（保过）、房产违规等9类约560词。
| Argument | Type | Description |
|---|---|---|
| mask | boolean | 是否返回打码替换后的安全文案，默认 True |
| text* | string | 待检测的广告文案全文 |
{
"type": "object",
"required": [
"text"
],
"properties": {
"mask": {
"type": "boolean",
"default": true,
"description": "是否返回打码替换后的安全文案，默认 True"
},
"text": {
"type": "string",
"description": "待检测的广告文案全文"
}
},
"additionalProperties": false
}We read every tool's name, description and inputs the way the model will, looking for instructions hidden from you. A clean result means no warning signs, not a guarantee. Scanned Nov 9, 2026.
No readOnlyHint / destructiveHint annotations, so clients can't tell a read from a write without asking.
Fix: Add annotations to each tool, e.g. { readOnlyHint: true } for tools that only read.