npm
profanease
观察池 · 暂无正式排名
该项目当前未满足“两个有效维度 + 两种数据源”的主榜门槛。
— 未排名
可观测采用度当前有效 · 2026-09-12 0
动量缺失 —
关注度缺失 —
信号可信度 依据当前有数据的独立评分维度数量计算。
有限1/3 · 1 种数据源
项目介绍
Modern, lightweight profanity detection & content moderation toolkit for JavaScript/TypeScript.
Profanease ships an MCP (Model Context Protocol) server, allowing AI assistants like Claude, Cursor, and others to use profanity filtering as a tool.
Claude Desktop (claudedesktopconfig.json):
Claude Code (.mcp.json in project root):
All tools accept: text, language (en/all), normalize, categories, customwords, customonly.
User: "Check if this user comment is appropriate: [text]" > The AI calls profaneaseanalyze with the text and gets back match details, severity, categories, and cleaned text — then makes an informed moderation decision.
Use Profanease.custom() to create a filter with only…
各数据源
598 周下载量
- 周下载量 598