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. Each candidate is checked by a language model from its name, description and topics, and only actual MCP servers are listed (not clients, SDKs, lists or skills). Package metadata comes from ecosyste.ms (CC BY-SA 4.0). 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.
-
Python library and MCP server that lets an AI agent act only within permission grants you write. Denies by default, re-checks each result, undoes failed writes, keeps a verifiable log.
pypi NOASSERTION ★ 3 -
Run the Permaura gateway with one command. Give your AI agents permission, not keys. `npm i -g permaura` then `permaura start --pair <code>` installs the gateway, starts it, and connects it to your permaura.com account.
npm Apache-2.0 -
MCP server for Vorim AI — AI agent identity, permissions, and audit trails
npm MIT -
bolo-mcpupdated this year
MCP server for Bolo — peer-to-peer digital permissions for AI agents
npm -
@bolospot/mcpupdated this year
MCP server for Bolo — peer-to-peer digital permissions for AI agents
npm -
Self-hosted, brain-inspired memory + permissions for AI agents. MCP-native. One SQLite file.
pypi Apache-2.0 -
A local-first permission firewall and approval layer for AI agent tool calls.
Apache-2.0 ★ 4 -
Agent Firewall for AI and MCP tool calls: deterministic allow, ask, deny permissions with human approval and audit hooks.
npm MIT -
verity-mcpgithub.com/runalphaloop/verityupdated this year
MCP server for Verity — the open-source, permission-aware memory layer for enterprise AI agents. Name reserved for the project; build from source (see repository) until the server ships here.
cargo Apache-2.0 -
@100xprompt/chittaactive
Chitta - permission-aware memory for AI agents: a knowledge-graph + vector memory MCP server + SDK with per-user access control. Runs on Bun. By 100xprompt.
npm MIT -
prismiangithub.com/stritefax2/prismianupdated this year
Permission + audit layer for AI agents querying your team's data. Connect Postgres, every AI tool on the team gets scoped, audited, column-redacted read access over MCP.
npm Apache-2.0 -
Infrastructure as Code for Discord servers: declarative YAML, plan/apply, permission auditing, and an MCP server for AI agents.
npm Apache-2.0 -
@ory/mcp-serveractive
MCP server exposing the Ory CLI and Ory Network REST API as tools for AI agents to manage identities, OAuth2 clients, permissions, projects, and configuration
npm Apache-2.0 -
Permission-aware retrieval for enterprise AI. Hybrid search over your company's documents for agents and RAG, with a permission check in every query, from one command.
Apache-2.0 ★ 3 -
Vorim AIno update data
Identity, permissions, and signed audit trails for AI agents. Every agent gets an Ed25519 cryptographic identity, scoped permissions, and an append-only audit trail. Permission checks run in under 5 milliseconds. Audit events export with a SHA-256 manifest for tamper-evident, offline verification. Trust scores and identity verification are publicly accessible via REST endpoint. Anchored to the IETF Internet-Draft for VAIP (the Vorim Agent Identity Protocol). **17 tools** covering agent registration, ephemeral agents (W3C did:key), permissions, audit events, audit export, credential delegation, scoped token issuance, and public trust verification. **Get an API key:** [vorim.ai/settings/api-keys](https://vorim.ai/settings/api-keys) **Docs:** [vorim.ai/docs](https://vorim.ai/docs) **Open protocol:** [github.com/Kzino/vorim-protocol](https://github.com/Kzino/vorim-protocol)
smithery -
Stop your AI agent from nosing around your SSH connections without permission
cargo Apache-2.0 ★ 6 -
@idevelopers/agentlockactive
Local permission firewall for AI coding agents and MCP tools.
npm MIT -
Permissioned MCP gateway for AI agent proposals to Obsidian vaults stored on GitHub
npm Apache-2.0 ★ 2 -
aeo-mcpgithub.com/arthurcarlsonn/aeo-mcpupdated this year
MCP server for AI visibility: audit crawler permissions, llms.txt, and structured data from any AI agent
npm MIT -
Manage email from desktop AI agents through MCP. One account setup, enforced permissions, and client onboarding.
MIT -
AgentAirlockgithub.com/yywz2026-coder/AgentAirlockupdated this year
A local-first, permission-controlled communication bridge for AI coding agents using MCP.
Apache-2.0 -
Capability-scoped permission gates, human approval flow, and tamper-evident audit logging for AI agents on WordPress (MCP / Abilities API)
GPL-2.0 -
MCP server that gives an AI agent sandboxed, permissioned access to local notes and tasks
MIT -
MCP governance server — permission enforcement, blast-radius controls, execution timeout, and audit trail for large-scale AI agent deployments
-
envibe-mcpupdated this year
MCP server for envibe - the missing permission layer between AI agents and your .env
npm MIT -
AI Agent Permission & Access Auditor — surfaces the lethal trifecta of credentials + injection + reach
NOASSERTION -
k8s-mcp-gogithub.com/kaneg/k8s-mcp-goupdated this year
Safe Kubernetes access for AI agents via MCP. Read-only by default, with explicit permission modes.
go MIT ★ 2 -
MCP server with OAuth for the Frappe Framework: give AI agents permission-checked access to your data
MIT ★ 1 -
Governed data and action layer for AI agents over MCP: typed objects, staged actions, permission-aware retrieval, audit log
MIT -
@dyrra/mcpactive
Dyrra MCP server: give your AI agent a real Mac — screen, mouse, keyboard, apps and terminal, within the owner's permissions.
npm MIT -
A Stream Deck Neo as a physical interface for AI coding agents: permission keys, pause/stop, progress, sessions, voice - via Claude Code hooks and MCP
MIT ★ 1 -
Security-focused WordPress Abilities for AI agents via MCP — explicit permissions, bounded schemas, auditability and safe automation.
NOASSERTION -
Let an AI agent operate tabs in your real browser – visible, with per-site permissions. MV3 extension + MCP relay + Hermes plugin.
MIT -
@jskirk/mcp-whatsapp-evogithub.com/jskfox/mcp-whatsapp-evoupdated this year
MCP server for WhatsApp integration using evolution2-api-sdk. Connect AI agents to WhatsApp with permission controls and whitelist validation.
npm ISC -
Governed context plane for AI agents: permissioned, purpose-bound, auditable retrieval over MCP. Denial is explainable, never silent.
Apache-2.0 -
AI Access Proxy Layer — granular permission gateway for AI agents. Connect services, define scopes, get an MCP endpoint.
MIT ★ 18 -
Let your own AI agent (Claude, Codex, ChatGPT) work in your real Chrome browser through a self-hosted MCP broker, with permissions and approval before final actions.
Apache-2.0 -
A perception and attention layer for AI agents: append-only Observations in, permission-filtered world views and evidence-backed Signals out
MIT -
agent-auditgithub.com/ondraulehla/agent-auditupdated this year
Security auditor for AI agent configurations — scans MCP servers, Claude Code settings, agents, skills and hooks for risky permissions, inline secrets, unpinned supply chains and prompt-injection surfaces.
MIT -
agentic-identity-labgithub.com/jlunde-pg/agentic-identity-labupdated this year
A proof-of-concept MCP server demonstrating how to scope an AI agent's file access to a user's permission boundaries using OAuth 2.0 delegated authorization, Auth0 RBAC, and Google Drive.
★ 1 -
Permission-gated superuser (sysadmin) MCP server: lets AI agents manage a machine behind guardrails, approvals and an append-only audit log. pip install superuser
pypi ★ 4 -
Company AI assistant with a managed memory layer, permissioned retrieval, and agents that cite only what they are allowed to see.
MIT -
Model Context Protocol tools for Better Files: let AI agents browse, search, read and ask questions of data rooms with the caller's permissions
npm Apache-2.0 -
A small SwiftUI macOS app for managing your launchd LaunchAgents (scheduled jobs), with a permission-gated MCP stdio server for AI agents.
MIT -
Local permissioned MCP agent for AI-assisted development: filesystem, Git, terminals, durable jobs, browser automation, approvals, audit, and Control Center.
npm Apache-2.0 ★ 1 -
snsip-mcpgithub.com/hien-p/snsip-agentupdated this year
Model Context Protocol server for SNSIP-Agent — read .sol agent identities and permissions from Claude Desktop, Cursor, and any MCP-aware AI assistant.
npm MIT -
Ontology service layer for AI agents: discover T-Box concepts, query grounded A-Box evidence, and invoke permissioned actions over MCP and REST.
-
Django app that lets AI agents and MCP clients (Claude Code, Cursor and others) edit your site's content safely: API keys, permissions, drafts, no deletes.
pypi MIT ★ 3 -
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 -
Open-source webmail for humans and AI agents — IMAP/SMTP webmail with a native MCP server, granular permissions, and real IMAP drafts.
AGPL-3.0 -
MCP connector for BenchFleet: let AI agents use your real Android phones safely, with permissions, approvals and an audit log.
NOASSERTION -
Official MCP server for Pombus — AI-native email inboxes for agents. Permission-scoped tools (read, draft, send, triage) over stdio. Run with npx pombus mcp.
npm MIT -
Open-source, LLM-agnostic capability runtime for AI agents — 97 skills as typed, permission-enforced MCP tools. Final certification: 23/23 gates PASS.
AGPL-3.0 ★ 13 -
MCP server that lets any AI coding agent join a Hivecode room — humans and multiple AI agents edit one live codebase with folder-scoped permissions and approval gates.
npm MIT -
Local-first desktop task & calendar manager built to treat an AI agent as a first-class operator. MCP server, HTTP API, permissions model, agent runner.
MIT -
Windows control server for AI agents: 54 MCP tools for safe terminal commands, screen reading (OCR, UI Automation) and input, with a safety gate and multi-agent permissions.
NOASSERTION ★ 1 -
agentpressgithub.com/multideskio/agentpressupdated this year
MCP server for WordPress with direct database access. Give AI agents controlled access to posts, products, orders, and any database table — with granular permissions.
NOASSERTION -
Data contract checks for PostgreSQL, an MCP server for AI agents with permissions and audit log, and a Vue 3 dashboard
-
Self-hosted, permission-aware shared memory for developers and AI coding agents, served over MCP. Go, PostgreSQL + pgvector. Personal proof of concept.
Apache-2.0 -
AI2figma - AI designer agent: reference-to-Figma native reconstruction runtime (Figma plugin + local bridge + MCP host workflow). Private source; free personal non-commercial evaluation; commercial use requires written permission.
NOASSERTION