Server facts
chinese-sensitive-words-mcp
Project
- Author
- CCCpan
- License
- Not specified
- Category
- AI Models & Agents
Tech
- Runtime / language
- Not specified
- Transports
- stdio
- Remote available
- No
- Authentication
- Not specified
Activity
- GitHub stars
- 104
- Last commit
- Data as of
Tags
- Claude
- chinese
- compliance
- content moderation
- mcp
- nlp
- openclaw
- qclaw
- sensitive words
Overview
Chinese social publishing involves platform-specific wording restrictions, from exaggerated advertising claims to prohibited livestream language. Chinese-sensitive-words-mcp brings checks for Xiaohongshu, Douyin, Kuaishou, and Bilibili into an MCP workflow. The check_sensitive_words tool examines submitted text and reports flagged terms with their categories, locations, severity, and suggested substitutions where available. Results distinguish high, medium, and low risk from advisory notices, helping writers identify which passages need attention. Its dictionaries cover subjects including political sensitivity, sexual content, violence, gambling, drugs, advertising superlatives, and medical efficacy claims.
Detection also addresses disguised wording: homophonic substitutions, symbols, and characters inserted between letters or words. Phone numbers and links are recognized, while named-entity filtering aims to reduce false positives involving people, places, and organizations. After reviewing the findings, writers can use get_word_suggestions to look up alternatives for a particular keyword or retrieve the replacement library grouped by category. This supports both checking a complete draft and revising individual expressions. The service connects through stdio, and each checking request supports up to 3000 characters.
chinese-sensitive-words-mcp setup notes
- chinese-sensitive-words-mcp records stdio only; confirm a network transport upstream before planning remote access.
- chinese-sensitive-words-mcp has no recorded license; confirm reuse terms in its repository.