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.
-
relay-mcpgithub.com/arclat-ai/relay-mcpupdated this year
Relay - open-source AI agent latency auditor. A local MCP server + CLI that finds, verifies, and fixes latency bottlenecks in AI agent and voice-agent codebases. No API key, no backend.
MIT ★ 2 -
Model Context Protocol (MCP) server for LLM spend & latency intelligence. Ask Claude why our LLM bill jumped.
★ 3 -
relay-arclatgithub.com/arclat-ai/relay-mcpupdated this year
Relay - AI agent latency auditor as an MCP server and CLI (an Arclat product)
pypi MIT -
ALPNAI tools for authorized AI agents: deterministic cost, latency and quality analysis, reproducible examples, API/MCP integration and report delivery.
NOASSERTION -
probeops-mcp-servergithub.com/kumarprobeops/probeops-mcp-serverupdated this year
MCP server for infrastructure diagnostics and geo-proxy browsing. SSL, DNS, uptime, latency, traceroute, port scan, and geo-located web browsing from 6 global regions.
npm MIT -
An agent-friendly, open-loop load tester for HTTP and gRPC, in Go. Honest latency, and it tells you when the generator is the problem. CLI, Go package and MCP server.
MIT -
model-radar-mcpgithub.com/srclight/model-radarupdated this year
MCP server that discovers live catalogs across HTTPS providers and subscription CLIs, ranks models by latency, and helps AI agents pick or compare them.
pypi MIT -
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 -
Isolythgithub.com/adarshthakur9240/Isolythupdated this year
Load-tested Isolyth's MCP tool-calling endpoint with 100 concurrent virtual users via Locust, sustaining ~535 req/sec at p95 latency of 160ms with 0% error rate; benchmarked WASM-sandboxed code execution against plain tool calls, showing only ~15-30ms of added latency overhead from sandboxing —proving security isolation came at min performance cost
-
Observability mcp-server for Lockwood Group — hard p99 latency budget under 150ms
-
Siglagithub.com/apkd/Siglaactive
Low-latency code search MCP server for Rust, C# and Unity projects
MIT ★ 2 -
pgrgithub.com/entireio/pgrupdated this year
PGR is an experimental, stateless MCP code-search server for studying how ranking, latency, and output shaping affect agentic search.
MIT ★ 63 -
Design-time sustainability co-pilot (MCP server): greener, cheaper cloud infrastructure decisions with carbon + cost + latency tradeoffs.
pypi MIT -
ReasonForge-MCP-Servergithub.com/RoyCoding8/ReasonForge-MCP-Serverupdated this year
A deterministic tool-calling framework for small LLMs. Integrates SymPy and sandboxed Python to achieve better accuracy/latency without massive parameter counts.
Apache-2.0 -
Industrial-grade, local-first memory bunker for AI agents. 9.7MB Rust binary, sub-millisecond latency, zero data leaks.
GPL-3.0 -
edge-diagnostics-x402github.com/seob5285-alt/edge-diagnostics-x402updated this year
Edge-based network diagnostic utility checking host latency, gated via x402 payment rails for ai agents.
MIT -
A fast, low-latency Rust-based Model Context Protocol (MCP) server querying local rustdoc JSON via SQLite FTS5 for AI agents.
MIT ★ 1 -
Compare OpenTelemetry traces across releases, gate latency and error regressions in CI, and inspect the evidence through MCP. Built with Kotlin and SQLite.
MIT ★ 1 -
synapse-mt5github.com/mfalmeh/mcp-mql5updated this year
Synapse MT5 — Native C++ MCP server bridging AI assistants to MetaTrader 5. Sub-millisecond latency, zero Python dependencies.
npm SEE LICENSE IN LICENSE -
Read-only MCP server for VMware ESXi: CPU Ready, disk latency, snapshots, real space usage, hardware health and events
pypi MIT -
multimodal_search_mcp_architecturegithub.com/estelacode/multimodal_search_mcp_architectureupdated this year
Low-Latency Multimodal Search Web App with MCP Architecture 🔌🛠️🧠 #StrandsAgentSDK #FasterWhisper #Multilingual #MultimodalSearch #ChromaDB # DockerCompose
★ 1 -
Official Model Context Protocol (MCP) server for Kovitel Vietnam eSIM, carrier APN rescue, and 5G latency benchmarks.
MIT -
MCP server for safe programmatic access to the local Tailscale daemon: tailnet discovery, SSH config generation, port sharing via Serve/Funnel, and latency matrices.
npm CC-BY-NC-ND-4.0 -
freshprobegithub.com/Sudhan30/freshprobeupdated this year
Data freshness verification for AI agents. Probes endpoints for cache staleness, latency, TLS health, and content changes. CLI + MCP server + HTTP API.
MIT ★ 2 -
LLM cost calculator, token counter, latency benchmark, CI guardrail, MCP server, and VS Code/Cursor extension.
MIT ★ 1 -
api-health-mcpgithub.com/marilynceo/api-health-mcpupdated this year
API uptime and performance monitoring for AI agents. Check endpoints, measure latency, validate responses, verify SSL certificates, and track SLA compliance u2014 all from your MCP client.
-
MCP server for safe programmatic access to the local Tailscale daemon: tailnet discovery, SSH config generation, port sharing via Serve/Funnel, and latency matrices.
NOASSERTION ★ 1 -
Scalable MCP backend optimized for Claude Code and Claude Desktop — secure, low-latency, plug-and-play.
MIT ★ 3 -
ContextHubgithub.com/BryceEWatson/ContextHubupdated this year
Deterministic MCP server for coordinating AI agents across software development lifecycles. SQLite graph traversal, Zod validation gates, RFC 6902 patching, and 100+ tests with <200ms p95 latency.
-
Deterministic in-process AST security runtime, secret zeroization, and Ed25519 Merkle receipts for Cursor, Claude Code, and autonomous AI agents (<35µs latency).
pypi Apache-2.0 -
High-Trust Open-Core framework for Windows AI Agents. Crushes latency to 3.9ms, enables background UIA invocation, and protects enterprise RPA workflows.
MIT ★ 3 -
Sovereign local-first AI voice studio. Neural speech synthesis, zero-shot voice cloning, zero-latency dictation, and MCP agent voice I/O. Built with Tauri & FastAPI.
AGPL-3.0 -
@probeops/mcp-servergithub.com/kumarprobeops/probeops-mcp-serverupdated this year
MCP server for multi-region infrastructure diagnostics — SSL, DNS, ping, whois, traceroute, port check, latency, and geo-proxy browsing from 6 global regions
npm MIT -
pipecat-docs-mcpgithub.com/DTufail/pipecat-docs-mcpupdated this year
MCP server for Pipecat voice AI framework documentation. Search 6,936 chunks from 317 pages via Claude Desktop or Cursor. Hybrid BM25 + FAISS retrieval with cross-encoder reranking. 78.8% MRR, 93% Recall@5, 36ms avg latency.
-
Universal AI Co-Producer for DAWs · Bit-transparent ITU-R BS.1770-4 telemetry, 128-band RTA & zero-latency native accessibility bridges (macOS / Logic Pro & Steinberg).
NOASSERTION -
DIOgithub.com/nisaral/DIOactive
Drop-in OpenAI- and Ollama-compatible LLM gateway that learns each backend's latency online and routes vLLM / SGLang / TGI / Ollama with SLO-aware admission. No engine patches.
Apache-2.0 ★ 3 -
sentinel-aigithub.com/MaxwellCalkin/sentinel-aiupdated this year
Real-time AI safety guardrails for LLM apps. 10 scanners: prompt injection, PII, harmful content, code vulnerabilities, obfuscation detection. Sub-ms latency. Python + TypeScript SDKs. MCP proxy. Claude Code hooks.
Apache-2.0 ★ 27 -
mcp-network-toolsgithub.com/rog0x/mcp-network-toolsupdated this year
MCP server for network diagnostics: DNS lookup (A/MX/TXT/NS), IP geolocation, SSL certificate checker, WHOIS lookup, HTTP ping with latency stats Install: npx @rog0x/mcp-network-tools
NOASSERTION ★ 1 -
The open routing layer for AI tools: an MCP server that routes agent tool calls (search, extract, browse) across providers per policy — cost, latency, pinning, deny lists — with failover and the routing decision on every result. BYOK.
npm BUSL-1.1 -
NeuralBridge_mcpgithub.com/dondetir/NeuralBridge_mcpupdated this year
🧠 AI-native Android automation via MCP. Give Claude, GPT, Gemini or any AI agent full control over Android — tap, swipe, type, screenshot — with ~6ms latency. 100x faster than Appium. 43 tools. No root required.
NOASSERTION ★ 48 -
Experimental, pre-release. Drives Android devices and emulators, iOS Simulators and iPhones for AI agents: a daemon with a live mirror of every screen, a CLI and an MCP server, built around the latency of each tap-and-observe step
MIT ★ 6 -
Hosted streamable HTTP MCP server for Cloud World Model: simulate cloud infrastructure and estimate cost, latency and resilience before deploying or changing infrastructure. Multi-cloud world model for AWS, GCP, Azure, OCI and DigitalOcean; MCP and REST, with x402 payment on eligible requests.
-
weather-mcp-a2agithub.com/Ratnesh-181998/weather-mcp-a2aupdated this year
MCP-based agentic AI system enabling LLMs to fetch and reason over real-time global weather data using standardized tool invocation. Built with MCP Server/Client, Llama 3, and external APIs to deliver low-latency, hallucination-free weather intelligence through protocol-driven workflows.
MIT ★ 1 -
Free proxy list, verified around the clock: HTTP, HTTPS, SOCKS4 and SOCKS5 proxies with country, anonymity, latency and uptime on every row, updated several times a day. Live list, proxy checker, keyless API and MCP server at hproxy.com.
MIT ★ 92 -
The open routing layer for AI tools. MCP server that routes agent tool calls (search, extract, browse) across 8 providers per policy — cost, latency, pinning, deny lists — with failover and the routing decision on every result. One Go binary, BYOK, no account. Source-available, BUSL 1.1 → Apache 2.0.
npm NOASSERTION ★ 2 -
Nornicdb is a distributed low-latency, Graph+Vector, Temporal MVCC with all sub-ms HNSW search, graph traversal, and writes. Using Neo4j Bolt/Cypher and qdrant's gRPC means you can switch with no changes while adding intelligent features like schemas, managed embeddings, reranking+llm, GPU accel, Auto-TLP, Policy-based Memory Decay, and MCP server.
MIT ★ 931 -
Eyes, ears and hands for AI agents on GUI programs: a sealed X11 or Wayland session where an agent (over MCP), a script or CI reads the accessibility tree, takes screenshots, records video and sound, clicks, types and drives gamepads, then checks contrast, Tab order, visual diffs, flicker, input latency and audio dropouts. CLI, Python, Playwright.
npm MIT ★ 1 -
token-scoutgithub.com/jackccrawford/token-scoutupdated this year
For OpenClaw, Hermes and more. Find free and low-cost inference (LLM models). Use them directly. Provides both a CLI and MCP server that knows which free-tier LLM APIs exist, which ones you have keys for, and which one fits your task. Returns endpoints so can you call models directly. No proxy, no middleware, no latency tax.
MIT ★ 9 -
Graph-Augmented Code Intelligence RAG engine pairing deterministic AST call graphs with dense vector search to boost direct-callee Recall@8 from 0.0% to 53.3% (+53.3 pp) under a 5.8ms p95 latency bound. Powered by a QLoRA fine-tuned Qwen2.5 adapter (70.02 CodeBLEU), quality verification harness, and production Model Context Protocol (MCP) server.
MIT -
Self-hosted Bitbucket for AI agents, the gap Atlassian's Cloud-only Model Context Protocol ("MCP") server leaves open, plus a leaner Jira and Confluence at 60 to 78% fewer tokens than raw REST, same latency. 58 tools on FastMCP over stdio, or one Auth0 OAuth 2.1 connector URL for the whole org, every call audit-logged: person, AI client, tool
GPL-3.0 -
scrcpy-mcpgithub.com/JuanCF/scrcpy-mcpno update data
scrcpy-mcp gives AI agents real vision and control over Android devices. It exposes screenshots, taps, swipes, text input, key events, UI inspection, and arbitrary ADB shell commands as MCP tools, so an agent can drive a phone or tablet the way a person would. When a scrcpy session is active, input and screen capture run 10–50× faster than plain ADB, making real-time automation, on-device testing, and UI agent workflows practical. Highlights - Fast screenshots and live screen capture - Tap, swipe, drag, long-press, scroll, and text input - UI hierarchy dump and element search by text, resource-id, or class - Arbitrary ADB shell command execution - Optional scrcpy session for low-latency control Works with any ADB-reachable Android device (USB or Wi-Fi). Configure ADB_PATH, ANDROID_SERIAL, and scrcpy paths as needed.
smithery -
时间锚定的人机动作交接:agent 让人类在指定时刻执行物理动作,按压时间戳精确到毫秒(MCP 服务 + 桌面 cue 面板)
MIT ★ 1 -
MCP server: which AI model is up, fast, popular or low-priced right now. Public data, no API key. Node 18, zero deps.
MIT -
Configurable tile-based system monitor for Windows: per-CPU, processes, GPU, kernel (DPC/ISR per driver, pool tags, objects), an hour of history, plugins and an MCP server for AI agents.
MIT ★ 1 -
See every hop of a request through your local LLM system: a tracing server, a live flow map, a waterfall per request, and an MCP server for Claude Code and Cursor. One static Go binary.
Apache-2.0 ★ 1 -
Task Manager on steroids: a Windows task manager that stays responsive under load and shows who is doing what, and why. Kernel tracing, stutters, network, history, and an MCP server for AI assistants.
Apache-2.0 ★ 19 -
Fast computer use for AI agents: event-driven settling MCP server - replaces time.sleep between clicks, waits for the UI to react and stop changing, returns the settled screenshot. No blind waits, no wasted vision turns.
NOASSERTION -
Rust + WASM sublinear-time solver for asymmetric diagonally dominant systems. Exposes Neumann series, push, and hybrid random-walk algorithms with npm/npx CLI and Flow-Nexus HTTP streaming for swarm cost propagation and verification.
npm MIT ★ 94 -
Fast deterministic MCP pre-filter that blocks known scope-creep (S3) and prompt-injection (S2b) tool-call patterns before an agent runs them. ~25 microsecond overhead, no LLM in the hot path; a first-line pre-filter, not a sandbox.
npm Apache-2.0