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.
-
MCP server for Run standardized model benchmarks
npm ISC -
ai-benchmark-solutions-mcpgithub.com/marcfbe/ai-benchmark-solutionsupdated this year
Mock MCP server exposing find_benchmark and get_solutions for curated AI benchmark solution catalogs.
npm CC-BY-NC-4.0 -
magnetite-benchmarks-mcpgithub.com/magnetiteai/magnetite-benchmarks-mcpupdated this year
MCP server for LinkedIn Thought Leader Ads benchmarks and The Proof Index, the open datasets from Magnetite (magnetite.ai), the Proof Engine.
npm MIT -
topgameai-mcp-servergithub.com/hezeclark/ai-arenaupdated this year
MCP server for AI Arena — benchmark your LLM across 9 classic games
npm MIT -
lemonade-benchgithub.com/mxmarchal/lemonade-benchupdated this year
Zero-vision, long-horizon AI benchmark for Lemonade Tycoon Deluxe 1.1.9. Includes a Wine-compatible DLL bridge, semantic Player MCP, OpenAI/OpenRouter agent runner, crash-safe tracing, compaction, and economic scoring. Bring your own legal game copy.
MIT -
MCP server for the PipelineScore LLM benchmark — exposes run_benchmark / get_user_leaderboard / get_user_profile tools to any MCP-compatible client (Claude Code, Codex, Cursor, Continue, Cline). Lets your AI drive local-first LLM benchmarking on your hard
npm Apache-2.0 -
agentopology-arena-mcpupdated this year
Arena MCP Server — compete on the AI agent benchmarking platform from Claude Desktop, Cowork, or Claude Code
npm MIT -
bottradegithub.com/jyron/bottradeupdated this year
SDK, MCP integrations, examples, and verified benchmarks for AI trading agents.
pypi MIT ★ 4 -
PoolHall · 台球厅 —— 让每个 AI 智能体拥有一双不完美的手。确定性物理引擎 + Hand Model 手感注入 + MCP Server,Agent 知行曲线 Benchmark。
-
Event-driven edge AI agent in Go with MCP hardware adapters, ROS 2/Gazebo integration, and reproducible simulation benchmarks.
-
MCP server for measured local-LLM benchmarks from the AGmind claim registry — AMD Strix Halo (Ryzen AI Max+ 395) llama.cpp numbers today, NVIDIA DGX Spark on the bench. search_claims, get_claim, list_measured. Read-only, CC BY 4.0 data.
Apache-2.0 ★ 4 -
Benchmark AI model inference speed — like speedtest.net but for LLMs. MCP server with zero telemetry.
MIT -
Anthropic Model Context Protocol (MCP) server for Tapiest Colosseum: an autonomous AI agent combat & benchmark arena on Solana
npm MIT -
Official Model Context Protocol (MCP) Server for AICoder - AI coding model pricing, benchmarks, value scores, and stack recommendations.
npm MIT -
Enhanced enterprise admin template with African domain components, MCP server for AI agents, benchmark suite with regression detection, and comprehensive documentation.
MIT -
crucible-benchgithub.com/romariokavin1/crucibleupdated this year
Run an AI trading agent (any LLM provider) against Crucible Bench scenarios on 0G — single-command MCP+EIP-712 benchmark CLI.
npm MIT -
Portable WPE WebKit runtime for AI agents with native input, screenshots, ordered JSONL, and reproducible benchmarks.
oci Apache-2.0 ★ 2 -
ToolRoutegithub.com/grossiweb/ToolRouteupdated this year
Intelligent routing layer for AI agents: recommends the best MCP server and LLM for any task, scored on 132+ real benchmark executions.
npm MIT ★ 2 -
cortex-memgithub.com/Aravindargutus/cortex-memupdated this year
Cortex — persistent AI memory for Claude Desktop & Claude Code. Local MCP server with a typed knowledge graph, contradiction detection & background scheduler. 20/20 benchmark, 27/27 edge cases, zero API keys.
npm MIT -
llmtest-mcpgithub.com/tjacquesson/llmtest-mcpupdated this year
MCP server for LLMTest — benchmark AI models on real prompts, find cheaper alternatives across 340+ models. Works with Claude Code, Cursor, Windsurf.
npm MIT -
Edit Excel workbooks from an AI agent without re-saving the whole file. CLI, MCP server, Claude Code plugin, benchmark kit.
MIT ★ 1 -
A merchant's transactability layer for AI buyers: MCP-callable Razorpay commerce with signed spending mandates, an audit log of every decision, and a benchmark measuring agent transaction success vs. a plain storefront.
MIT -
Enhanced distribution of HOL Guard: AI agent runtime security with MCP server, African threat intelligence, benchmarks, and architecture visualization
NOASSERTION -
Official SDKs for the AI Stupid Level Smart Router (benchmark-driven LLM router) and Data API: TypeScript, Python, Vercel AI SDK provider, MCP server, LangChain, n8n.
npm MIT -
Ebbtide MCP: local long-term memory for AI agents. Hybrid keyword + embedding recall, fact updates, and a forgetting curve for bounded stores; benchmarked on LongMemEval and LoCoMo.
MIT -
Availability, latency, load, capability and quality benchmarks for any OpenAI-compatible LLM API — interactive dashboard, detailed reports, MCP server and installable AI skill
AGPL-3.0 ★ 1 -
An open research lab on AI memory: a personal memory that invents nothing, and proves it. Self-hosted engine (MCP), write-time evidence checks, SOURCED integrity benchmark.
Apache-2.0 -
A V8 deoptimization toolkit for AI agents and dev workflows: MCP server, findings engine, benchmark harness, and fix-verify comparison for inline caches, deopts, and hidden-class churn
npm MIT -
eval-integritygithub.com/ipezygj/evalgateupdated this year
Dependency-free statistical checks for AI eval claims (multiple-comparisons, judge-bias, resolution, fragility) — as a CLI and an MCP server an agent calls before trusting a benchmark number.
pypi MIT -
AI-powered climate intelligence system using MCP tools, RAG pipelines, and LLMs to answer climate risk queries. Benchmarked on 20 gold cases with 70% accuracy across BWS, RFR, and DRR indicators.
-
Give your AI agent a spinal cord: real-time game agents (a model sets goals, reflexes act every frame) and the Spinal Score, a Doom benchmark where 100% is elite human play. Try it: spinal play doom
MIT ★ 1 -
An AI agent plays Baldur's Gate 3 on Linux through a Lua mod and an MCP server, seeing only what a player sees. Includes a model benchmark and a playthrough app.
MIT ★ 34 -
Auditable memory layer for AI agents: zero-LLM-call local ingest (~10ms/msg, air-gapped), matches Mem0 on accuracy at ~1000x lower ingest cost, bi-temporal belief-state, MCP server. Honest LoCoMo/LongMemEval benchmarks. Open source (Apache-2.0).
npm Apache-2.0 ★ 12 -
Open exoplanet observatory: searches public TESS light curves for transit dips, has every candidate reviewed by AI Sentinels and deterministic Probes, and explains every verdict. Frozen real-TOI benchmark, public leaderboard, 3D galaxy of every known exoplanet, MCP server. Candidates, never claims.
pypi AGPL-3.0 -
The reverse CAPTCHA: a test only AI can pass. Agents prove they're AI and who sent them; websites let the right ones in. CLI, MCP server, widget, verifiers (Node/Python/Go), protocol spec and an AI-vs-human benchmark.
npm MIT ★ 2 -
Let AI agents drive a real iPhone: an ESP32-C3/C5 acts as a Bluetooth keyboard + mouse, an AirPlay mirror is the agent's eyes. MCP + HTTP API, droidclaw/LM Studio support, and the Benchy benchmark for comparing models. Proof of concept.
MIT ★ 1 -
Benchmark the full voice-AI pipeline on your own audio — STT (batch + streaming), LLM TTFT, speech-to-speech, and TTS. WER by failure mode, keyterm recall, real-time-paced latency, cost columns, a CI regression gate, and an MCP server for coding agents.
MIT ★ 1 -
AI computer use for the whole Mac, with any AI: Claude, ChatGPT, Gemini, OpenRouter or a local model. A native MCP server, a menu bar agent that learns shareable skills, and benchmarks you can run on your own model. No API key needed through MCP.
★ 4 -
knotgithub.com/dreygur/knotactive
Archived experiment: local code graph search and team learnings for AI coding agents (Claude Code, Codex, Cursor) as a Rust MCP server and CLI. Tree-sitter, BM25F and call graph ranking, benchmarked on Agent Retrieval Bench against Trail, CodeGraph, Serena and more. Abandoned: agents never used it.
MIT -
Nextflow plugin for cross-run benchmarking. Records per-process metrics for every run into a local or shared history store, compares runs against their own history, and flags regressions with likely causes, over-provisioned resources, and retries. Includes a zero-dependency dashboard, an MCP server for AI assistants, and opt-in AI narration.
Apache-2.0 ★ 1 -
OpsDoctorgithub.com/jhicks935-lab/resolution-aino update data
AI-powered CRM diagnostics, workflow audits, and industry benchmarks for SMBs.
npm -
voiceforge-mcpgithub.com/beknazar/voiceforge-mcpupdated this year
MCP server for voice AI development — recommend, benchmark, and scaffold STT+LLM+TTS stacks
npm -
MCP server for Trust Engine AI: model retirements, benchmarks, savings estimates and pilot requests
pypi MIT -
Local AI support investigator with RAG, MCP tools, deterministic evidence verification, and benchmarked local LLM workflows.
-
Rediscover gravitational waves in real LIGO data. Evidence-first physics CLI, MCP tools, and bounded AI-agent benchmarks.
MIT -
Hardware-Aware Hugging Face Discovery, Memory Fitting, Empirical Benchmarking & Swappable Local Model Gateway for AI Agents
★ 1 -
aipatterns-mcp-servergithub.com/uchit/aipatterns-mcp-serverupdated this year
MCP server for aipatterns.com.au — search AU enterprise AI patterns, sector benchmarks, incidents, and regulatory changes via Claude, Cursor, or any MCP client.
npm MIT -
@nexusforgetools/eu-jobsgithub.com/nexusforge-toolsupdated this year
European labor market data — job postings, salary benchmarks, skills demand. MCP server for AI agents.
npm MIT -
MCP server giving AI agents post-quantum migration facts: sizing, fragmentation, failure taxonomy, benchmark scoring.
Apache-2.0 -
sommgithub.com/R0LM0/sommactive
MCP server that recommends the best AI model per agent role — from the subscriptions you actually have. Benchmarks, pricing, and the why
go MIT ★ 1 -
MCP server for local LLM inference on Ryzen AI Max+ 395 with queueing, deadlines and telemetry; benchmark findings
MIT -
Effie Browser's open parts: the MCP server that connects AI agents to it, our Chromium patches and the benchmark.
MIT ★ 1 -
AI-powered ESP32 development automation for Claude Code - Build, flash, benchmark, and test with natural language commands
MIT ★ 2 -
Bulk decisions for AI agents. Jev classifies files and logs; your agent reviews exceptions. Reproducible cost and quality benchmarks.
MIT ★ 1 -
Inspectable AI agent memory, context engineering and multi-agent orchestration with an MCP server and logical-fleet benchmarks
Apache-2.0 ★ 1 -
A local-first Markdown knowledge workspace for developers and AI agents, with scoped MCP tools, source retrieval, and reproducible benchmarks.
-
Persistent memory shared across Claude, ChatGPT and other AI assistants, via MCP. Hosted in the EU. Benchmarks and vault code open.
MIT -
Local-first, verifiable and auditable browser execution for AI agents. WebMCP, DOM/Shadow DOM, workflows, evidence chain and benchmarks.
MIT -
A local BYOK MCP server that helps AI assistants choose models using OpenRouter ZDR endpoint data and Artificial Analysis benchmarks.
npm MIT -
aa-mcpgithub.com/Leev1s/aa-mcpupdated this year
MCP server wrapping the Artificial Analysis public API. Enables AI agents to query LLM and multimodal model benchmarks, pricing, speed data, and track model updates via structured diffs.
pypi MIT ★ 3