Wanhe AI Compliance Checker
Offline MCP server that checks Chinese ad copy for banned terms and AI-content labeling
- Pricing
- Open source
- Platforms
- CLI, self-hosted
- Version
- 1.1.5
- Listing updated
- today
Similar MCP servers
About Wanhe AI Compliance Checker
Wanhe AI Compliance Checker is an open-source MCP server for security. It is available on CLI and self-hosted.
The README describes a Model Context Protocol server that lets AI assistants such as Claude Desktop, Cursor and Cline check Chinese commercial writing for advertising-law problems before publication. It targets product pages, short-video scripts, WeChat articles and AI-generated text, returning each flagged phrase with its position, the legal basis cited and a suggested rewrite.
The server exposes three tools: a banned-term check, a test for whether AI-generated content needs explicit or implicit labelling under the Chinese labelling measures, and a 17-item organisational compliance questionnaire. The README says the free version runs locally with a rule subset, while a cloud version with the full rule library is described as planned or pending a filing.
The README says it is aimed at Chinese marketing and content teams and their developers, and states that its output is rule matching and not legal advice.
Tools this server provides (3)
Taken from the project documentation; the running server may differ.
- check_content
- Checks text for compliance risks and returns each issue with its basis and suggestion.
- check_ai_label
- Determines whether content needs explicit or implicit AI labelling.
- self_assessment
- Scores answers to a 17-item organisational compliance questionnaire.
What it does
- Banned term detection
- Flags absolute claims, medical efficacy wording and false promises in Chinese text, citing the relevant article of the Advertising Law.
- AI content label check
- Assesses whether text generated by AI needs explicit or implicit labelling under the stated labelling measures.
- Organisational self-assessment
- Runs a 17-question compliance questionnaire and returns a coverage score with a list of items to fix.
- Offline local operation
- The free version runs on the user's machine without network requests, according to the README.
- Legal basis per finding
- Each issue is returned with its matched text, position, cited basis and a suggested edit.
Good for
- Reviewing Chinese e-commerce product pages for banned advertising claims before publishing
- Checking short-video scripts for medical or absolute wording
- Deciding whether AI-generated images or text need a visible AI label
- Giving an AI agent a compliance check step while drafting Chinese marketing copy
Pricing
Open source
Free local version under MIT licence; a cloud version is described but its availability is pending.
Prices change. Check wh-bi.github.io before you buy.
Wanhe AI Compliance Checker FAQ
Does it send my content to a server?
The README says the free version runs entirely locally, collects no usage data and makes no network requests unless the user configures the cloud version.
How is it installed?
The README gives a git clone and a self-test command, plus a Claude Desktop configuration example using a stdio command and arguments. It states that no pip install is required.
Is it legal advice?
No. The README states that the tool performs rule matching and legal citation and does not constitute legal opinion.
Is Wanhe AI Compliance Checker free?
Yes. Wanhe AI Compliance Checker is open source, so you can use and self-host it for free.
What are the best alternatives to Wanhe AI Compliance Checker?
The closest alternatives to Wanhe AI Compliance Checker are Wanhe Compliance — Chinese Content Compliance Checker, 中国 AI 合规与备案(条文级自查), ComplyEdge TrustLint: offline AI law checks, Eu Ai Act and SynomosAI Governance MCP.
Where this comes from
- Description: written from the project's README, the project's website and its MCP Registry entry.
