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.
-
Hypothesis-first experiment tracking for agent-driven ML research, wired into Claude Code.
pypi MIT ★ 3 -
agentic-experimentsgithub.com/kadenmc/agentic-experimentsupdated this year
Git-first, hypothesis-forcing experiment tracking for agent-driven ML research. Bundles a research harness for the H->E->F artifact model, uses signac for local execution/run state, and bridges to W&B for remote observability.
pypi MIT -
Model Context Protocol (MCP) server for Comet ML experiment tracking: list, search, and analyze experiments from your AI assistant.
pypi Apache-2.0 ★ 2 -
Modern experiment tracking that follows you across servers, messengers, and LLMs.
pypi NOASSERTION ★ 29 -
MCP server for tracking ML experiments (metrics, hyperparameters) with SQLite
-
Experiment tracking built for comparison: multi-run charts, targets, baselines, tags and notes in one place. Bring experiments together across your LAN and let AI agents inspect them through a built-in MCP endpoint.
MIT -
Agent-native ML experiment tracking: write contract, code snapshots, knowledge graph, MCP server, live UI
Apache-2.0 -
Probe Research: experiment tracking for coding agents. Plugins and skills for Claude Code, Codex and Cursor plus a read-only MCP server, so agents build on what your team already tried. CLI: pip install probe-research / npx probe-research. CI-generated mirror of Probe's monorepo.
go Apache-2.0 -
uxon-aigithub.com/alexpilotto/uxon-aiupdated this year
MCP server and Claude skills for PPC landing pages, A/B experiments, and first-party conversion tracking. Build Google Ads and Meta pages, run cross-domain experiments, and pull CRO reports via the UXON API.
MIT ★ 11 -
skore-mcpgithub.com/fanfcorp/skore-mcpupdated this year
MCP server for Skore (experiment tracking & model evaluation)
pypi MIT -
MCP (Model Context Protocol) server support for trackio experiment tracking
MIT ★ 3 -
Self-hosted personal life-tracking dashboard — daily habits, Oura, training, bloodwork, goals & experiments. FastAPI + SQLite + HTMX, with a read-only REST API & MCP server.
MIT -
claude-labbookgithub.com/anthonylee991/claude-labbookupdated this year
MCP server — persistent experiment tracking and semantic code search for AI coding agents
npm Apache-2.0 ★ 15 -
mlops-mcp-servergithub.com/elliottdevo8/mlops-mcp-serverupdated this year
MCP server for MLOps workflows - experiment tracking, model registry, and deployment across MLflow, W&B, and SageMaker
MIT -
RL & control-aware MCP memory system for structured experiment tracking, theory audit, and safe pattern reuse.
MIT -
🔬 Enhance your experiments with SciAgent, an AI-powered tool for automatic tracking, intelligent analysis, and local storage—all without altering your code.
MIT ★ 1 -
MCP server for travel compliance and trip planning — visa requirements, tax residency, Schengen tracking, accommodation, transport, and experiences across 189 European destinations.
MIT -
Fathom Analyticsno update data
Access your Fathom Analytics data effortlessly through an AI assistant. Retrieve account details, manage sites, and generate insightful reports with real-time visitor tracking. Enhance your analytics experience with flexible tools designed for seamless integration.
smithery -
self-evolving CLI for DSPy program management — featuring hot-swap inference, MCP agent interoperability, a Generative Feedback Loop (GFL) pipeline, skills system, llms.txt generation, and full MLflow experiment tracking.
★ 1 -
Self-hosted portfolio journaling and agent-driven stock research: reproducible experiments, backtesting, durable jobs, API/CLI/MCP extensions, and multi-currency tracking.
MIT ★ 47 -
Track your whole life in seconds a day, then run experiments on yourself. Self-hosted, open source, AI-ready (MCP). Symptoms, sleep, mood, habits, money.
AGPL-3.0 ★ 1 -
QuantLab Pro is a quantitative ML research platform featuring six specialized agents, MCP-based tool integration, leakage-resistant time-series modeling, PyTorch experiments, backtesting, and reproducible MLOps tracking using Python, Streamlit, and SQLite.
MIT -
ExpHub is an open-source MCP server that transforms AI assistants into collaborative knowledge workers. It automatically searches past experiences, captures new solutions in structured templates, and tracks todo items — forming a continuously evolving knowledge network across all your AI tools.
MIT ★ 1 -
experiments-mcpgithub.com/Prateek-Gaurav7296/experiments-mcpupdated this year
Description: Query your MLflow experiments in plain English using Claude Desktop — MCP server built with FastMCP
MIT -
openclawmcpgithub.com/nanyun818/openclawmcpupdated this year
OpenCLAW MCP lab agent for AI-assisted training experiments.
MIT -
tempcastgithub.com/thanakara/tempcastupdated this year
Deep learning experiment station for weather time-series forecasting, powered by Visual Crossing API data.
-
chronohorngithub.com/asuramaya/chronohornupdated this year
Family-agnostic experiment tracker and architecture-search runtime, driven by a 64-tool MCP surface.
MIT ★ 3 -
Universal experiment provenance and reproducibility for science and engineering.
npm Apache-2.0 -
Experiment memory for research agents: register attempts, capture outcomes, and report reproducibility through MCP.
-
clearml-codex-plugingithub.com/SergeiNikolenko/clearml-codex-pluginupdated this year
Read-only ClearML MCP plugin for Codex with task search, metrics, logs, comparisons, and experiment snapshots.
MIT ★ 1 -
resmangithub.com/hanmahong5-arch/resmanupdated this year
Local-first experiment tracker for autonomous AI training agents - the verifiable memory layer for agent training loops. One Rust binary, zero config, MCP-native.
MIT -
Backtest-integrity auditor — is your edge real or overfit? Deflated Sharpe, PBO, cost break-even, one Robustness Score. Experiment tracker with hash-chained ledger + live drift monitor. MCP-native, 396 tests.
MIT -
Long-term debugging memory for AI coding assistants — prevents retry loops, tracks competing hypotheses, records redacted evidence, and verifies acceptance criteria.
MIT -
Local-first agent memory for research: literature evidence, experiment runs, and failure cases with provenance contracts, quantified effectiveness metrics, and a read-only MCP server for Codex, Claude Code, and Cursor.
-
Auditable research provenance as local-first MCP tools. The claim checker, paper feed, knowledge graph and symbolic math assume nothing about your field — any results JSON or CSV works, from any instrument. The experiment ledger adds W&B/MLflow/Sacred/DVC/Hydra for computational work. Nothing leaves the machine.
MIT ★ 1 -
Your training runs, live, on your own machines. A self-hosted ML run tracker: phone alerts when a run dies, every GPU in view, MCP for coding agents.
Apache-2.0 ★ 1 -
Agent-driven ML researcher in a secure, reproducible KVM lab: five standalone scientific-state MCP servers within per-user VMs with driver / trusted / hostile zone isolation
Apache-2.0 ★ 8 -
mcp-memory-suitegithub.com/sscctv/mcp-memory-suiteupdated this year
Comprehensive MCP memory plugin — AI coding memory, personal life memory (habits, experiences, knowledge, accounts), scheduled review, and cross-module linking
MIT