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.
-
ClinicalTrials.gov & OpenFDA Biotech / Pharma Lead Enricher ($0.80/1k) — Includes 1-Click n8n Workflow, Make.com Blueprint & Claude MCP Server (https://apify.com/lucacelli/clinical-trials-fda-drug-safety-intel)
-
MCP server for session continuity and sandbox safety in LLM development workflows. Provides "pick up exactly where you left off" + "never break the main app" functionality.
MIT -
osha-mcpgithub.com/dtjohnson83/osha-mcpupdated this year
MCP server for OSHA safety and compliance reference workflows.
★ 1 -
VibeLigngithub.com/yesonsys03-web/VibeLignupdated this year
AI coding safety CLI for vibe coding workflows. Checkpoints, undo, anchors, MCP, and secret protection for Claude Code, Cursor, Codex, and OpenCode.
MIT ★ 26 -
MCP server for n8n administration: workflow backup, rollback, duplication, node-param patching, and replay-sample-payload - with a pre-duplication write-safety policy
MIT -
Unofficial safety-focused MCP kernel for Kalshi market workflows
oci Apache-2.0 -
saferx-companiongithub.com/aldian/saferx-companionupdated this year
Medication safety concierge agent — ADK 2.0 workflow graph, custom openFDA MCP server, PII redaction, and HITL caregiver approval gate
MIT -
Trust and safety MCP server for AI agents. Detect grooming, bullying, self-harm, CSAM, fraud and deepfakes across text, voice, image and video, then triage a moderation queue and produce KOSA, DSA and GDPR evidence. Around 80 tools plus workflow skills for Claude Code and Cowork. OAuth sign-in, no API key to configure.
MIT -
oswaldgithub.com/austinAbraham/oswaldupdated this year
Oswald the Analytical Octopus — a runtime-agnostic, MCP-native, context-rot-resistant workflow layer for analytical-engineering AI agents. Eight tentacles, durable .oswald artifacts, safety policy engine, dbt build/test/validate. MIT.
npm MIT -
Python MCP server with 32 Spotify tools, OAuth authentication, multi-client integration, and safety-aware agent workflows.
★ 1 -
Safety-first MCP server and public gateway for licensed AutoCAD drawing inspection and plotting workflows.
MIT -
coding-mcpgithub.com/kieutrongthien/coding-mcpupdated this year
Coding MCP is a multi-project MCP server for remote coding workflows. It lets agents work across repositories without cloning source code or setting up each repo locally. It provides safety-first file, Git, and command automation, with HTTP/STDIO transports, auth, logging, and telemetry for production use.
MIT ★ 2 -
MCP-based AI interface for structured, simulation-first PX4 and robotic fleet workflows with deterministic safety checks
★ 15 -
A Telegram Bot control server for AI agents and automation workflows with proactive dispatch and zero-ban safety. 專為 AI 代理設計的 Telegram Bot 控制協議伺服器,支援主動派送、零風控與檔案傳發。
pypi MIT ★ 4 -
Nipmod package intelligence, Codex plugin, MCP workflows and public agent safety tooling.
MIT -
Native macOS system cleaner with YAML-driven safety rules, local AI explainability via MLX, and an MCP server for agent-controlled cleanup workflows.
AGPL-3.0 ★ 79 -
An MCP server, CLI, and Claude Code hook that validate GitHub Actions workflows for correctness and supply-chain safety.
MIT -
Claude Code plugin for Fly.io Sprites: computers for agents. Sprites MCP access, workflow skills, commands, and safety hooks for persistent remote development environments.
MIT ★ 4 -
Turns vague requests into structured prompts, execution plans, and policy-aware workflows — with readiness verdicts, PR merge-safety checks, agent packs, and MCP tool exports. Offline-first, deterministic, provider-agnostic.
Apache-2.0 ★ 29 -
github-ops-mcpactive
MCP server that lets your AI assistant work on GitHub: review pull requests, manage issues, branches, files, releases, workflows and security alerts, with built-in safety rules.
npm SEE LICENSE IN LICENSE -
MCP server for EasyEDA Pro, connecting AI agents to schematic and PCB workflows, controlled design operations, DRC, BOM sourcing, board analysis, and manufacturing export—with profile-gated tools and safety controls.
npm AGPL-3.0 ★ 72 -
A Model Context Protocol server that lets an AI agent drive a mock e-signature workflow, with a **human approval gate** on anything that sends mail or cancels a document. In-memory backend, no accounts or API keys, so you can learn MCP and agent safety without risk.
MIT -
A production-grade, self-hostable MCP server for Quera, bringing Quera’s problemset, submissions, courses, contests, LMS, AI, judge tooling, and authoring workflows to LLM agents. Built with async Python, 259+ tools, RBAC, safety gates, audit logs, revisions, multi-account support, and Docker deployment.
MIT -
Agentic-BIM-Compliancegithub.com/Prajwalkadam29/Agentic-BIM-Complianceupdated this year
This project replaces error-prone, 2-day manual structural reviews with a 4-minute Agentic AI workflow that parses IFC models for code compliance. It ensures structural safety through real-time 3D visual overlays, explainable reports, and automated "auto-fix" design suggestions.
MIT ★ 1 -
newrelic-mcpgithub.com/karldane/newrelic-mcpno update data
A Go-native MCP (Model Context Protocol) server for the New Relic observability platform, providing comprehensive access to APM, infrastructure, logs, traces, alerts, dashboards, synthetics, workflows, notifications, service levels, tags, entities, and workloads — all through a single GraphQL endpoint. Overview 55 registered MCP tools covering the full NerdGraph API surface Safety-first design: every tool carries EnforcerProfile metadata (risk, impact, resource cost, PII flag, idempotency) for automated policy enforcement via MCP Bridge Write protection: all mutation tools gated behind a -write-enabled flag; disabled by default Multi-region: US and EU data center support (NEWRELIC_REGION=us|eu)
smithery -
Approval workflows for AI agents
npm MIT ★ 33 -
Governed persistent AI workers with default-deny authority, human approvals, dry-runs, and an authenticated MCP.
Apache-2.0 -
Local-first workflow compiler that turns repeated AI-agent work into evidence-backed, quality-checked Skills.
MIT ★ 2 -
Local-first verification layer for AI agents: evidence, provenance, policy gates, human approval, and auditable Trust Receipts before a claim, memory write, or action is trusted.
npm Apache-2.0 ★ 31 -
midasgithub.com/omarkhandji-commits/midasupdated this year
Self-hosted AI agent for safe LLM workflows: approvals, receipts, budget controls, local dashboard, CLI, Ollama, and MCP.
MIT ★ 1 -
GAFF - Graphical Agentic Flow Framework: Complete MCP server suite for AI agent orchestration (meta-package)
npm MIT -
oracle-h-mcpgithub.com/tponscr-debug/oracle-h-mcpupdated this year
MCP Server for ORACLE-H — Human-on-the-Loop validation for autonomous AI agents
npm MIT -
Human approval requests for agent actions: expiring, single-use decisions. Trusted executors must enforce the gate; not a same-user security boundary.
npm MIT -
Synthetic telecom operations MCP server for agent workflows and safe tool-use experiments
NOASSERTION -
Portable MCP server for AI agents: persistent session checkpoints, resume and Total Recall, with package-aware build tools. Linux, macOS, Windows and SSH.
NOASSERTION -
contract-review-mcpgithub.com/tarunag10/contract-review-mcpupdated this year
Read-only MCP server patterns and examples for contract review workflows.
MIT -
SSH for AI agents (Claude Code, Codex, Cursor) where reads run freely and every change needs a signed human approval that the server itself enforces.
Apache-2.0 -
mcp-legal-toolkitgithub.com/tarunag10/mcp-legal-toolkitupdated this year
MCP tools and examples for safe, local-first legal and compliance workflows.
MIT -
Autonomous Security Guardrails & Governed Remediation for AI-Built Apps. Tri-Mode Parity (CLI, Chat Slash Commands, Native MCP Server), First-Principles Security Suite, Multi-Modal Vision OCR, 86 Rules across 22 Families & Ponytail Patch Bounds.
npm MIT ★ 1 -
Agentic workflow replay & audit with OTel GenAI semantic conventions
NOASSERTION -
sealgithub.com/useseal/sealupdated this year
Human-in-the-loop as infrastructure. An MCP-native gateway that pauses risky agent tool calls until a human approves — declarative policy, fail-closed by default, tamper-evident audit trail.
Apache-2.0 ★ 1 -
print-conciergegithub.com/dpdev69/print-conciergeupdated this year
Safe AI print concierge for Bambuddy with MCP workflows and human-confirmed queueing
★ 8 -
Safe AI operations agent demonstrating MCP tools, deterministic ERP workflows, approval gates, evaluations, and idempotent writes.
MIT -
flowcheckgithub.com/backslash-ux/flowcheckupdated this year
A Git hygiene safety layer for AI-first development. MCP server that monitors repositories and nudges AI agents to maintain clean commit history.
MIT -
agent-security-mcpgithub.com/mdfifty50-boop/agent-security-mcpupdated this year
MCP server providing security scanning, prompt injection detection, secret leak detection, and agent permission auditing for AI agent workflows
npm -
nagoc-sender-mcpupdated this year
Model Context Protocol (MCP) server for Sender.net REST API v2 with human-in-the-loop safety guardrails
npm MIT -
Portfolio extract of a production accounting-ops agent: closes the books over 10 typed MCP tools, auto-books only what a deterministic in-code policy gate proves safe, routes the rest to a human, prints a full audit trail. Every hazard branch is pinned by a stdlib check in CI.
MIT -
Vigilancegithub.com/B-A-M-N/Vigilanceupdated this year
A security minded , realtime AI Intent Observability MCP providing intent tagging and visual mirroring of AI attention to see what context the AI is looking at. Made for visual and coding workflows.
NOASSERTION ★ 1 -
claude-code-invoice-guardgithub.com/mikko-lab/claude-code-invoice-guardupdated this year
Deterministic guardrail (refuse, don't guess) for an LLM agent on a VAT-critical invoice workflow: skill + subagent + MCP server + PreToolUse hook. Portfolio/demo plugin.
MIT -
Limetry is open-source agent action governance: check tool intents against policy before irreversible runs. Returns allow, deny, or approval_required with signed receipts and privacy-safe audit. TypeScript SDK, CLI, and MCP for IDE like Cursor, VS Code, GitHub Copilot, Claude. Optional CI, SQL, and Shopify adapters. MIT — self-host your own node.
npm MIT ★ 1 -
Official ModelBound plugin for Cursor — slash commands for skill pipeline, trust & safety, versions, and MCP-backed context sync.
npm NOASSERTION -
Light Remote MCP turns a Windows or Linux machine into an explicitly authorized remote execution target for ChatGPT, Claude, Codex, and other MCP-capable AI workflows.
Apache-2.0 -
semantic-gps-hackathongithub.com/mboss37/semantic-gps-hackathonupdated this year
Mission control for AI agents. Governance gateway between AI agents and MCP-connected business systems: live policies, saga rollback, audit on every call, Tool Relationship workflow discovery. Built with Claude Opus 4.7.
MIT ★ 1 -
maishagithub.com/WinterLabsHQ/maishaupdated this year
Free & open-source MISRA C / CERT C / BARR-C compliance for embedded C — the vendor-neutral alternative to Polyspace, Helix QAC & Coverity's paid workflow. Agentic fix loop + MISRA Compliance:2020 audit evidence, from any IDE via MCP.
NOASSERTION ★ 8 -
Scenario-driven framework for evaluating demographic bias in single- and multi-agent LLM workflows. Tracks bias across sequential stages, separates model non-determinism from systematic effects, and prevents agents inferring evaluation intent. REST API and MCP server.
MIT -
sofagent is an open-source FDE Harness for AI coding agents — on entry, write your business judgment into files (workflows, ontology, AI nodes); after departure, audit every change against them. 28 codified audit rules, tamper-evident chain, snapshots, and an MCP toolset (104 tools, 12 plugins). Everything can be FDEing.
npm MIT ★ 56