Best mcp servers for ai & language models.
85 mcp servers in AI & language models. The top 20 are ranked by public GitHub stars, collected November 20, 2026; the rest are listed below them. Model gateways, inference, agent frameworks, memory, evaluation and observability for LLM applications.
| # | Listing | Category | Stars | Forks | Last push |
|---|---|---|---|---|---|
| 1 | Agent MCP Agent-MCP is a framework for creating multi-agent systems that enables coordinated, efficient AI collaboration through the Model Context Protocol (MCP). | AI & language models | 1.3k | 184 | Mar 28, 2026 |
| 2 | Comet Opik Model Context Protocol (MCP) server for Opik, the open-source LLM observability and evaluation platform, built by Comet. | AI & language models | 226 | 37 | Nov 18, 2026 |
| 3 | Synap Memory Persistent memory for AI agents — log and recall conversation context over MCP. | AI & language models | 81 | 18 | Nov 19, 2026 |
| 4 | Cairn.ink Memory Cross-session memory for AI agents with a Source Receipt for every memory, over MCP. | AI & language models | 56 | 6 | Nov 19, 2026 |
| 5 | daidocs Open plain-text file format for AI memory. | AI & language models | 55 | 4 | Oct 7, 2026 |
| 6 | Slowave Local adaptive memory for AI coding agents through a five-tool MCP lifecycle. | AI & language models | 26 | 2 | Nov 14, 2026 |
| 7 | Baron Munchausen Local grounded memory for coding agents: MCP over stdio, stdlib only, a verdict per answer | AI & language models | 12 | 0 | Sep 11, 2026 |
| 8 | Lockstep Decision memory for AI coding agents: captures decisions once, briefs every agent before it acts. | AI & language models | 12 | 0 | Nov 12, 2026 |
| 9 | novgraph A live knowledge graph of your codebase for coding agents: commit intent, co-change coupling, ranked blast radius, computed architecture - over MCP or the shell, at 78-99.74% token savings measured per answer. | AI & language models | 9 | 1 | Sep 26, 2026 |
| 10 | Causari Records what an AI agent did in a repo (prompt, model, files) and answers why a line exists. | AI & language models | 9 | 0 | Nov 16, 2026 |
| 11 | Blue Synergy Cortex Long-term, drift-free cognitive memory engine for AI Agents and Cursor IDE. | AI & language models | 8 | 1 | Nov 5, 2026 |
| 12 | picsio-mcp-plugin Pics.io MCP Plugin for AI Marketplaces (Anthropic, OpenAI, Copilot, Gemini) | AI & language models | 7 | 0 | Oct 30, 2026 |
| 13 | Magnemo Governed memory for AI agents. | AI & language models | 4 | 2 | Nov 9, 2026 |
| 14 | OpenReflex Local execution memory for AI coding agents: reuses past fixes and flags failure loops. | AI & language models | 4 | 0 | Nov 15, 2026 |
| 15 | al-buddy-memory Portable, governed, model-agnostic memory for AI agents: facts invalidated never deleted, raw text as truth, one documented export format. | AI & language models | 3 | 1 | Nov 19, 2026 |
| 16 | mechapad-cli Open Multi-Model AI Control Surface & Native MCP Server — Run Claude Fable 5.1, GPT-6 Astra, DeepSeek V4.1 Flash, and Gemini 3.8 in parallel | AI & language models | 2 | 0 | Sep 11, 2026 |
| 17 | MMW Managed Memory Workspace Ultra-fast persistent memory workspace with tenant isolation and cross-model continuity. | AI & language models | 2 | 0 | Nov 11, 2026 |
| 18 | astermind-mcp AsterMind MCP server - on-device RAG token reduction (stdio, 10 tools) | AI & language models | 1 | 1 | Sep 25, 2026 |
| 19 | Assertion Memory Recall your team's coding-agent memory. | AI & language models | 1 | 0 | Oct 7, 2026 |
| 20 | codex-reset-mcp Hosted read-only MCP server for OpenAI Codex usage-limit resets: 24/48h forecast, verified reset record, service status. | AI & language models | 1 | 0 | Sep 24, 2026 |
More mcp servers for ai & language models 65
Listed without a public GitHub repository, or not yet collected.
AcqPath — RSL rights preflight for AI & RAG
RSL rights preflight for AI, RAG indexing and training.
Agent World
Live data + LLM inference for AI agents: prices, uptime, x402 audits, chat.
AI Model Watch
Prices, context windows, lifecycle and deprecation dates for 270+ AI/LLM models.
AICB
AIContextBuilder (aicb) - Roslyn-backed context for coding agents and LLMs: MCP server, CLI and Windows desktop app for C#/.NET
alphai-agent-inspect-evaluation
Local evaluation fixture: agent-inspect 6.29.1 capture and retry-evidence check on a synthetic AlphAI MCP flow
AskAiRank
How ChatGPT, Claude, Perplexity and Gemini mention your brand: visibility, mentions, competitors.
AuditAE
Check how ChatGPT, Perplexity, Gemini and AI Overviews cite a brand, and fix it on any website.
biohacking-kompakt
Biohacking Kompakt für Claude und Gemini: Evidenz-Check zu Supplements, Peptiden und Biohacking mit BK-Score (Plugin + Gemini-CLI-Erweiterung, Remote-MCP)
Bluet
Product memory your agents read before they work: briefs, plan checks against decisions, debriefs.
Bunstation
Build websites and apps from a prompt, edit files, publish, and check credits and usage.
Contexel
Shared AI memory. ChatGPT, Claude, Cursor and any MCP app read and write the same memory.
contextcut-mcp
Local-first AST code context pruner for LLMs & AI coding agents (Claude, Cursor, Antigravity) via MCP.
ContextDB Customer Memory
Customer memory with recall, confirmation, forgetting, and pre-action evidence checks.
contextetech
Hub open source de ressources pour LLM : contextes, prompts, datasets, LoRA, serveurs MCP, agents, skills, évaluations, RAG — contextetech.com
cubite-mcp
MCP server for Cubite LMS, a hosted learning management system (SCORM, xAPI, certificates, cohorts).
Engram
Memory for AI agent teams across tools, sessions, repositories, and teammates.
First Dollar Test
Base agent evaluation: 2 USDC fee, 1 USDC on verified PASS, full fee refund on FAIL/expiry.
Free Agentic Publication Digester
Cited daily digests of official US federal publications.
GardenArena GardenScript
Simulate farms, validate GardenScript and compare policies with Rust.
gather
Attestable memory for AI agents: curated writes, verifiable tenant isolation, bi-temporal facts.
Genesis402 (x402 pay-per-call)
179 x402 pay-per-call APIs: OpenAI-style chat, embeddings, web extract, wallet/token briefs
genpark-prompt-injection-jailbreak-neutralizer-skill
Zero-dependency heuristic adversarial prompt injection & override neutralizer (Lakera / NeMo)
HallucC — Hallucination & Safety Guardian
Hallucination & safety checks for LLM/Agent outputs: claim-level fact-check with citations.
HeatPulse
AI-news gateway for agents: curated events, digests, live LLM leaderboards, intent router.
Questions and answers
What is the most-starred mcp server for ai & language models?
Agent MCP, with 1,309 public GitHub stars as of November 20, 2026. Agent-MCP is a framework for creating multi-agent systems that enables coordinated, efficient AI collaboration through the Model Context Protocol (MCP).
How many mcp servers for ai & language models are listed?
85 on RUAGENTIC, 20 of them ranked by stars and 65 listed without a public GitHub repository or without collected metrics.
How is this list ranked?
By public GitHub star count, read hourly from the repository each listing links to, last collected November 20, 2026. Stars measure attention on GitHub, not quality or safety. Listings that share one repository appear once. Sponsorship never changes a rank.
See every mcp server at /servers, the full mcp servers leaderboard for ai & language models, or list your project.