MCP servers tagged llm
10K live MCP servers tagged "llm", ranked by trust score. Tags come from package metadata and repository topics, so this list covers servers across every category that work with llm.
ThesisDefender Kaggle
ThesisDefender is an adversarial multi-agent AI system that stress-tests arguments, exposes hidden assumptions, evaluates resilience, and generates stronger revisions using Google…
Mathsketch Tutor
Sketch a math problem, get step-by-step feedback grounded in symbolic verification. NestJS + Firebase + mathjs CAS verifier + LLM exposed as an MCP tool.
Omakase
Omakase (omks) — orchestrate your installed AI agent CLIs (Claude Code, Codex, Gemini, Cursor) with Dynamic Workflows and a Goal-loop.
Skill Md
Define an AI agent in one file and deploy it as a pay-per-call REST + MCP endpoint. The SKILL.md spec + examples, used by Plinth (plinthai.xyz).
Amber Protocol
Amber Protocol: repository-local governance for coding agents, including a DeepSeek Harness (dsh) patch overlay.
Open Dungeon
Open source AI Dungeon
Sentricore Agent Security
Security gateway for AI agents on REST and MCP: prompt-injection defense, tool governance, human-in-the-loop escalation, tamper-evident audit logs.
Causal Field
Gaussian resource-contention fields for multi-agent coordination
Tri Party Framework
Orchestrate multi-agent workflows using Codex, Claude, and Gemini with verified cross-audits, source tracking, and structured merge gates to ensure output integrity.
ShodhRAG
Build a local-first AI assistant using retrieval-augmented generation, agents, and tool calling, powered by Rust for secure and fast performance.
Adaptive MCP Orchestrator Blueprint Showcase
Showcasing a Adaptive MCP Orchestrator Blueprint
Simdiff
The effect layer for AI-agent firewalls — simulate an action, get a canonical effect delta. Zero-dep, deterministic, fail-closed. 5 domains (fs, sql, shell, http, solana).
Cursor Chief Of Staff
Your AI chief of staff: persistent memory, task management, meeting notes, MCP integrations, daily Slack briefings, and a self-improving agent that gets smarter over time. Built o…
LLM Tool Registry
FastAPI backend for managing an AI-agent tool catalog — JWT auth, RBAC, encrypted secrets, remote MCP server probing, audit logging & rate limiting.
Castle Wyvern
A decentralized multi-agent AI infrastructure with Gargoyles theme. The Manhattan Clan is awake.
Pypkg Audit Ptc Agent
MCP-based context engineering patterns (PTC + PTD) achieving 51% token reduction vs ReAct for LLM agent systems
Ewa
Iterative understanding loop — fresh agent each round, graph as only memory, sema vocabulary
Qwen3.5 Fix
A high-performance proxy middleware to fix Qwen3.5 tool calling issues and enable seamless MCP client compatibility.
Productagent Evalbench
A local and optional-real-provider Agent evaluation bench for RAG, Memory, Skills, MCP-style tools, tracing, model benchmarking, and cost-controlled evaluation.
Mozaia Laws MCP
MCP server — programmatic access to the Mozambican law library for AI agents (Claude Desktop, Cursor, VS Code)
Mcpslim
Transparent MCP proxy that compresses tool outputs — cut Claude Code/Cursor token usage on the MCP tokens. Measured 90% on real list responses, 12-60% lossless-leaning by default.
Qwen Local Guide
Interactive, 100% local build guide: run Qwen on a Mac mini M4 (16GB) → RAG + multi-agent shared memory. One self-contained, offline HTML file.
Receipts
Catch AI-hallucinated code before you run it — checks every symbol against your installed packages.
Sector Rotation Agent
Macro-driven, multi-agent AI that ranks equity sectors by business-cycle regime — RAG over 20 years of analogs + Fed narrative, with an audit/critic safety layer.
Lore Builder
FastMCP RAG corpus builder
Claude AI Learning
Learning Claude AI with Anthropic & AWS Bedrock
N8n AI Builder
Build n8n workflows from natural language — works with Claude Code, Codex, Cursor, Windsurf, Copilot. Includes 8 expert skills, MCP server integration, validation, and push/pull s…
Agent Ready Starter
Make your site discoverable and usable by AI agents in ~30 minutes: llms.txt, OpenAPI spec, and an MCP server from one config file.
Trace Store
Append-only JSONL store, one durable record per MQO interaction: bind, grounding, execute, quality signals
ThreatModel
Open-source, MCP-first threat modeling that proposes and sandbox-validates the fix — not just the report. Apache-2.0."
Figma Build Design Skill
Agent skill for building Figma screens through your design system — library-first via Figma MCP, not manual redraws. Works with Claude, Codex & Cursor.
Code Chunk
AST-aware chunking of code for contextual retrieval into the Unison brain.
Mirror Mirror
Structured self-reports for long-running LLM sessions: operator signals, explicit uncertainty, no claims of model feelings.
Session Continuity Kit
Multi-turn session management for agent systems. Conversation history windowing, token budget management, context compression, and persistence. Adapters for Firestore, DynamoDB, R…
FreeCADClaude
Unofficial FreeCAD 1.1 workbench that docks a Claude chat panel — drives your own Claude Code CLI to inspect, view, and edit the active document through curated tools. Personal us…
AI Agent Bridge
AI Agent Bridge for VS Code — local models, DeepSeek, Codex, and Claude with native tools, memory, context control, and diagnostics.
Fwh Anthropic
Facetwork domain: Anthropic API wrappers (Messages / Batch / Files / Agent SDK / Claude Code / Computer Use) + DocumentQA workflow.
Dvara Examples
Reference Docker Compose stacks, cloud deploy recipes, and SDK integration samples for the DVARA LLM Gateway
Autovault Website
AutoVault documentation and product website
Async AI MCP Automation Service
Production-grade, multi-tenant async Python pipeline for AI candidate evaluation, LLM orchestration, and Anthropic MCP integration.
Syrudas
Local-first AI workspace with pluggable model providers — chat, agent mode, MCP, local RAG, deep research, model arena, writing editor, and a hardware-aware model cookbook. FastAP…
Hec Ovi
Profile front page: Senior Applied AI Engineer, agentic engineering. Multi-agent systems, MCP and agent tooling, RAG/CAG accuracy, generative AI (video, image, audio, 3D), local i…
Family Hubs
Recall — an open long-term memory layer for AI agents. MCP + REST + dashboard on Cloudflare Workers, Supabase pgvector, Workers AI. Free tier.
Toolsweep
Which tool-schema decision is costing you accuracy? Controlled factorial ablation for LLM tool APIs.
ScrapeForge
A self-learning web scraper that uses LLMs to generate CSS extraction schemas once per page template, then caches and auto-generalizes URL patterns. Zero manual selectors. Zero ma…
Open Character Memory
Open-source role-user memory runtime for role-play and tutoring agents | 面向角色陪聊与 AI 教育的角色-用户双向记忆模块
Hello MCP Dotnet
A production-shaped hello-world Model Context Protocol server in .NET, with a console chat client driven by a small local LLM.
Khwan Client Ts
Khwan TypeScript client (npm)
Aitdp
AI Threat Detection Protocol — open spec, community ruleset and reference SDK for detecting prompt injection, jailbreaks, data exfiltration and tool abuse in LLM apps and AI agent…
Lattice MCP Py
Python policy gateway for agent tools. Shadow mode, approval, audit. Zero required deps.