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.
-
cedar-mcpgithub.com/BACH-AI-Tools/cedar-mcpupdated this year
—
MIT -
A Model Context Protocol (MCP) server for interacting with the CEDAR (Center for Expanded Data Annotation and Retrieval) metadata repository.
MIT -
MCP server exposing the CEDAR artifact library's builders, readers, and renderers as composable tools for LLM orchestration.
NOASSERTION -
cedar-mcp-servergithub.com/Pigius/cedar-mcp-serverupdated this year
MCP server for Cedar policy language - validate, authorize, diff, and plan policy changes for Amazon Verified Permissions from your AI assistant.
NOASSERTION ★ 8 -
Multi-agent travel assistant on Bedrock AgentCore — A2A, MCP, per-tool Cedar authorization, and a dual-judge evaluation harness.
MIT -
tidb_agentcore_gateway_mcp_toolsgithub.com/bernard-kavanagh/tidb_agentcore_gateway_mcp_toolsactive
Governed fact-of-record layer for AI agents — deterministic adjudication, authority policies, tamper-evident history and tenant isolation, on TiDB + Amazon Bedrock AgentCore + MCP + Cedar.
-
Alexa+ add-on (MCP server + MCP Apps) that acts for you, keeps a receipt for every action, and turns corrections into Cedar rules enforced by Amazon Bedrock AgentCore Policy.
MIT ★ 1 -
Asimov's Three Laws compiled to real Cedar authorization policy that acts as the runtime safety layer for AI agents which touch the physical world. Governs the real SO-101 (LeRobot) arm over MCP.
★ 1 -
Transparent API security discovery: risk scores that cite their reasons, real-traffic cross-referencing, an agent/MCP security graph, behavioral baselining, and Cedar policy generation. Rust core, Python CLI/SDK.
Apache-2.0 -
Runtime governance for AI agents: intercept agent-to-tool traffic, decide against a versioned Cedar policy bundle, and write a tamper-evident decision ledger a third party can verify offline.
npm Apache-2.0 ★ 2 -
MCP Pulse — stream live context to LLM agents via MCP, Kafka & WebSockets. Real-time anomaly detection + AI root-cause diagnosis, Cedar authZ, Bedrock/OpenAI/Ollama. Includes an auto-insurance edition. Fork it! 🍴
Apache-2.0 -
Multi-tenant control plane for S3-compatible object storage: tenant-scoped credentials, Cedar policies, quotas, audit and events. Not an S3 proxy — clients move bytes over presigned URLs. Connect API, web console, Go and Python SDKs, and an MCP server for AI agents.
go AGPL-3.0 ★ 3 -
Governed, self-improving multi-agent swarm for 5G network operations on AWS: Bedrock agents, MCP tools behind AgentCore Gateway, Cedar policies, human approval, RL + federated learning on a digital twin. Finalist, AWS x FYUZ Breaking Barriers 2026 (team project).
-
secure-rag-multi-agentgithub.com/ssdeanx/secure-rag-multi-agentupdated this year
Secure Retrieval-Augmented Generation (RAG) with role-based access control using Mastra AI orchestration, with JWT Secure, also Deep Research built with Cedar & Next.js. Gemini, OpenAI, Openrouter, & Gemini CLI usable. Also exposes tools, agents as MCP Server Also A2A Agent Cards
NOASSERTION ★ 19 -
Safe, auditable Kubernetes actions for AI agents: plan, policy-check, approve, apply.
Apache-2.0 -
The system of record for what an AI agent knew, was allowed, and did — signed Agent Decision Records on the MCP tool path.
pypi Other/Proprietary License -
Self-hosted credential broker for AI agents: agents ask for an API call by credential name; a policy engine decides; secrets never reach the agent.
AGPL-3.0 ★ 14 -
wardergithub.com/santosh-singhh/warderupdated this year
A policy gate for MCP tool calls. Blocks destructive AI agent actions before they run — as the subprocess your AI client launches, where network gateways can't see.
cargo MIT -
The transactional write gate for coding agents: Plan -> canonical Manifest -> set-level checks -> hash-gated two-phase apply. MCP server, CLI, PreToolUse hook, GitHub Action.
npm MIT -
Zero-trust gateway for AI agent tool calls: policy-checked, approval-gated, and recorded in post-quantum signed receipts anyone can verify offline.
MIT -
Open-source AI agent security gate: deterministic ALLOW/BLOCK/ESCALATE authorization before every tool call, plain-English policy compilation, tamper-evident audit ledger. MCP security gateway + Claude Code/Cursor guardrails. Rust, Apache-2.0.
Apache-2.0