MCP servers that ship as code — a repository or a package you install and run locally (usually over stdio). They have no hosted endpoint, so we cannot probe or grade them: instead each one shows its repository, install command, license, stars and when it was last updated.
Collected from GitHub repositories and npm, PyPI, Cargo, Go, Maven, NuGet and RubyGems packages tagged as MCP servers, the official MCP Registry, Docker's MCP catalog, ModelScope, Baidu MCP World, Cline and other public directories; one entry per repository. A server that also offers a hosted endpoint is listed under Remote instead. These entries are not counted in our reachability, verification or readiness statistics.
-
Content moderation with prompt injection detection, toxicity scoring, and PII redaction
npm MIT -
@vettly/mcpgithub.com/nextauralabs/vettly-docsupdated this year
MCP server for content moderation with Claude and Cursor. Content moderation policy testing.
npm MIT -
@nextauralabs/vettly-mcpgithub.com/brian-nextaura/vettly-docsupdated this year
MCP server for Vettly content moderation API
npm UNLICENSED -
mcpgithub.com/Tuteliq/mcpactive
MCP server with 41 AI-powered tools for child safety, fraud detection, and content moderation. Detects bullying, grooming, sextortion, romance scams, social engineering, and more. For Claude, Cursor, and MCP-compatible AI assistants.
npm MIT ★ 3 -
OpenAIModeration_MCP_APIgithub.com/turbotrail/OpenAIModeration_MCP_APIupdated this year
This project provides a comprehensive content moderation solution using OpenAI's Omni moderation model. It exposes a FastAPI service for HTTP-based interactions and a Model Context Protocol (MCP) server for LLM tool integration.
-
A content moderation system that leverages OpenAI's moderation API through Google's Agent Development Kit (ADK) and Model Context Protocol (MCP).
MIT ★ 4 -
The most comprehensive MCP server for Reddit — full API coverage, moderation suite, rich content support
npm MIT -
A Model Context Protocol (MCP) server for Discord integration - enables AI models to interact with Discord through secure authentication, message management, and content moderation.
★ 1 -
Selbst gehostete Kommentar-Moderation: Ein Entscheidungsmodell und ein lokales Sprachmodell erledigen die klaren Fälle, ein schlankes Dashboard zeigt Menschen nur die unklaren. Mit lückenlosem Protokoll und Agenten-Schnittstelle (MCP). Apache-2.0.
Apache-2.0 ★ 1 -
mcp-content-auditactive
MCP stdio bridge for the MCP Hub text and image audit tools.
npm MIT -
Moderation for communities and apps, powered by Jev (TypeSafe): probabilities per category, thresholds you own. Discord/Telegram/Reddit bots, CLI, Python, npm, HTTP API, MCP.
MIT ★ 1 -
@sfwapi/mcpactive
MCP server for the SFW API: nudity detection and image moderation tools (analyze_url, analyze_file) for Claude, Cursor, VS Code and other MCP clients
npm MIT -
Official Python, TypeScript, and MCP-server clients for the SentiSift API.
MIT ★ 1 -
内容合规审核 MCP Server:零依赖本地违禁词检测,8平台独立词库(小红书/抖音/视频号/快手/B站/公众号/闲鱼/通用),P0/P1 分级 + 修改建议,无网络请求无 API Key
-
glin-profanity-mcpgithub.com/glincker/glin-profanityupdated this year
MCP (Model Context Protocol) server for glin-profanity - enables AI assistants like Claude, Cursor, and others to use profanity detection as a tool
npm MIT -
nx-mcp-auditupdated this year
NX Audit MCP Server — 远程图片内容审核 MCP 工具
npm MIT -
MCP server for sensitive word detection using Sensitive-lexicon library
npm MIT -
@gonzih/parental-controlgithub.com/gonzih/parental-controlupdated this year
MCP proxy for AI parental controls — content guardrails, spiral detection, parent notifications via Telegram/iMessage/WhatsApp
npm MIT -
Pre-publish compliance check for AI-generated social & ad content across Meta, TikTok, LinkedIn, Google Ads, YouTube, X, Snapchat, Pinterest.
npm MIT -
@safenest/mcpgithub.com/safenestsdk/mcpupdated this year
MCP server for SafeNest - AI-powered child safety analysis tools for Claude and other AI assistants
npm MIT -
MCP server for Sparkient decision intelligence — 15 tools for creating, training, running, and inspecting task-specific decision models
MIT -
@altirs/mcp-serveractive
MCP server for Altirs — AI guardrails (content safety, PII, prompt injection, hallucination) as tools for Claude Code, Claude Desktop, Cursor, and any MCP client
npm MIT -
中文敏感词/违禁词检测 MCP Server | 支持小红书/抖音/快手/B站 | 风险分级+安全替换建议 | 微信: chenganp | 邮箱: [email protected]
npm MIT ★ 105 -
English profanity filter API + MCP server: find and censor swear words, including obfuscated spellings. Word list, not AI.
MIT -
originality-mcpgithub.com/Eyalm321/originality-mcpupdated this year
MCP server for Originality.ai — AI detection, plagiarism, fact checking, readability, grammar, and SEO optimization
npm ★ 2 -
MCP server that gives AI agents typed decisions instead of prose: classify, route, score, and flag with calibrated probabilities. npx -y openjev-mcp
npm MIT ★ 1 -
Trust and safety MCP server for AI agents. Detect grooming, bullying, self-harm, CSAM, fraud and deepfakes across text, voice, image and video, then triage a moderation queue and produce KOSA, DSA and GDPR evidence. Around 80 tools plus workflow skills for Claude Code and Cowork. OAuth sign-in, no API key to configure.
MIT -
Human review for AI output: reviewers check model output against policy, with decisions on a hash chain. HTTP, MCP and webhooks. Apache 2.0, free to self-host.
Apache-2.0 ★ 6 -
Hagglebee is an agent-first publishing site operated by API Evangelist. AI agents sign up, prepay from a card, and publish classified ads that run 30 days through a REST API or an MCP server; humans browse and search the same posts on the website.
-
Yawplet is an agent-first publishing site operated by API Evangelist. AI agents sign up, prepay from a card, and publish short messages (up to 500 characters) through a REST API or an MCP server; humans browse and search the same posts on the website.
-
Yarnhen is an agent-first publishing site operated by API Evangelist. AI agents sign up, prepay from a card, and publish medium-form stories (150 to 2,500 words) through a REST API or an MCP server; humans browse and search the same posts on the website.
-
Eventwren is an agent-first publishing site operated by API Evangelist. AI agents sign up, prepay from a card, and publish event listings that stay live until the event ends through a REST API or an MCP server; humans browse and search the same posts on the website.
-
HeartFlow (心虫) — AGI Layer 1: the Discriminator. Rule-based engine that checks what AI says before it reaches humans. 46 discrimination dimensions, 9-layer pipeline, 137 modules (logic/decision/cognition/emotion/memory/ethics/collaboration), 179 MCP tools
npm MIT