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-sentinel-scannergithub.com/j420/mcpsentinalupdated this year
MCP server for security scanning — analyze any MCP server against 177 detection rules with AST taint analysis, capability graph, and entropy-based secret detection
npm MIT -
Enhanced Binance MCP server with security fixes and technical analysis capabilities. Based on original work by @snjyor.
Apache-2.0 ★ 1 -
Security-first MCP server for TP-Link Omada, with a stdio-first runtime, capability-gated tools, and OpenAPI-based network automation.
npm MIT ★ 30 -
pentest-docker-mcpgithub.com/MaorOhana/pentest-docker-mcpupdated this year
Kali Linux-based MCP server for educational pentesting with ephemeral Docker containers. Integrates nmap, nikto, sqlmap, dirb, and searchsploit with Claude Code using isolated, short-lived containers for enhanced security. Direct connection approach with full Linux capabilities support.
★ 1 -
architect-mcp-servergithub.com/ageborn-dev/architect-mcp-serverupdated this year
A powerful, self-extending MCP server for dynamic AI tool orchestration. Features sandboxed JS execution, capability-based security, automated rate limiting, marketplace integration, and a built-in monitoring dashboard. Built for the Model Context Protocol (MCP).
MIT ★ 25 -
Policy-driven secure operations gateway for exposing bounded capabilities to AI agents, MCP clients and automation systems.
Apache-2.0 -
Capability-contract enforcement for AI agent tool calls — signed authority, a mediator that decides allow/deny/approve, and a tamper-evident ledger. Includes 3 reproducible incident labs and a conformance suite that declares its own uncovered paths.
Apache-2.0 -
Ephemora Cell — The execution layer for untrusted AI-generated code. Fast, capability-based WASM execution with explicit CPU, memory, time, I/O, and filesystem limits.
pypi BUSL-1.1 -
The scripting language of AI agents: the agent writes it, you review it at a glance, and the program can only do what it declares.
cargo Apache-2.0 ★ 3 -
Local-first compounding intelligence for AI agents. Experience Graph, Memory Bank, Growth Merge, learned skills. MCP-native for any model. MIT.
MIT -
Capability-based credential broker and MCP server for coding agents. Use approved APIs without injecting secrets into prompts, argv, env, or agent code.
MIT -
Local credential broker for AI agents. Keep API keys and SSH keys in an encrypted vault on your own machine; give each agent its own scoped, expiring, revocable grants; let it use a credential over MCP or ssh-agent without ever seeing it. Deny by default, every attempt in a hash-chained audit log. Single user, local only. v1.0.0.
go NOASSERTION ★ 1 -
Deny-by-default MCP server for GitLab: capability-based access for AI agents, a .noai content marker, and an authenticated git smart-HTTP proxy.
MIT ★ 1