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.
-
agentic-guard-mcpgithub.com/sanjaybk7/agentic-guard-mcpupdated this year
—
pypi NOASSERTION -
Deterministic AST Quality Gate, surgical code search & multi-stack verification MCP server (100% ML-free)
npm MIT -
Deterministic AST Quality Gate & multi-stack code verification MCP server (100% ML-free)
npm MIT -
Infraveil Agent Guardgithub.com/infraveilhq/infraveil-guardno update data
Gate an AI agent's destructive actions behind human approval, with a tamper-evident log.
pypi -
agent-guard-mcpgithub.com/dygogogo/agent-guard-mcpupdated this year
MCP Guard Server - Budget control, approval workflows and audit logging for AI agents (Claude Code, Cursor, ChatGPT)
MIT -
A safety layer for AI agents that spend money: payment gate, signed-approval verifier, MCP server and evals.
MIT -
agent-guard-mcpgithub.com/Kavachh/agent-guard-mcpupdated this year
Git-aware safe file operations MCP server for AI agents — delete anything except git-tracked source code. Safe to auto-approve.
mcpb MIT -
AI agent budget management & anomaly detection | x402 USDC | JPYC ready
-
Guard an AI agent before it acts: prompt-injection check for untrusted content + verify-before-pay for x402 endpoints and ERC-20s. Zero deps, MCP server, AgentKit actions.
npm MIT -
agent-guard-mcpgithub.com/mdfifty50-boop/agent-guard-mcpupdated this year
MCP server that detects and prevents infinite agent loops — circuit breakers, pattern detection, and stuck-agent recovery for agentic systems
npm -
Agent runtime guard — full-stack runtime interception for AI agents (Agent 运行时全栈拦截). A zero-dependency MCP server that blocks destructive commands, credential theft, SSRF and injection in real time, with Ed25519 evidence receipts. Works with Claude Code,
npm Elastic-2.0 -
Deterministic pre-execution safety gate for AI agents: verify tool calls against trusted evidence, policies, and WorldState before side effects.
★ 1 -
Let an AI agent write to Postgres one statement at a time: dry-run first, current-value checks, row cap enforced in the transaction, masked snapshots. CLI + MCP server, no dependencies.
MIT -
arc-agent-guardgithub.com/Mnorbert87/arc-agent-guardupdated this year
Give an AI agent a wallet it can spend from autonomously, but only within policy: limits, allowlist, approval threshold, audit trail. On Circle Arc, USDC as gas. MCP tools.
MIT -
MCP server that lets an LLM read chats, voice notes and images and draft replies, behind a tested send gate and a 15-case red-team eval.
MIT ★ 1 -
MCP server that fully parses and classifies shell commands (SAFE/BLOCKED/UNCERTAIN) before AI coding agents execute them, requiring a human-signed single-use token for anything uncertain.
npm MIT -
agent-guardgithub.com/sneiko/agent-guardupdated this year
Contract-based accountability runtime for AI agents. Define tasks with required evidence — no proof means not done. MCP server with multi-agent pipelines, independent verification, and append-only audit log.
MIT -
agent-output-guard-mcpgithub.com/agenson-tools/agent-output-guard-mcpupdated this year
Revolutionary MCP server that solves the 36.9% coordination failure rate in multi-agent systems. Zero LLM costs, pure computation. Built on UC Berkeley MAST study evidence.
npm MIT -
Lightweight security guardrails for tool-enabled agent runtimes: scan risky payloads, redact secrets, expose stdio/HTTP MCP tools, and write audit logs.
npm MIT -
Real-time streaming data contract gatekeeper & zero-trust Delta Lake quarantine engine. Features PySpark Structured Streaming, Kafka (KRaft), Model Context Protocol (MCP) shift-left validation, dbt/DuckDB analytics marts, and automated LLM incident triage handling 40K+ events/sec with sub-millisecond latency.
MIT -
CLI/MCP agent surface for Enji Guard
NOASSERTION ★ 4 -
skills-guard-mcpgithub.com/skills-guard/skills-guardupdated this year
Skills Guard MCP Server - Model Context Protocol Server for Agent Skills Security
npm MIT -
sh-guard-mcpgithub.com/aryanbhosale/sh-guardupdated this year
MCP server for sh-guard: semantic shell command safety classifier for AI coding agents (Cursor, Cline, Windsurf)
cargo GPL-3.0-only -
SAI Guard MCP server. Verifies transaction proposals for MCP-compatible agents; does not sign or broadcast.
npm MIT -
ux-guard-mcpgithub.com/x1n-q/ux-guardupdated this year
MCP (Model Context Protocol) server that exposes ux-guard as a tool for AI coding agents (Claude Desktop, Cursor, Blackbox Code, etc).
npm MIT -
@vulsight/guard-mcpactive
Advisory MCP server that asks Cerberus before an agent pays. For enforcement, use the proxy URL or the @vulsight/guard SDK hook.
npm MIT -
Agent 行为守护桌宠:跨 agent 监控可疑外传(加密打包直传对象存储 / DNS 外泄线索 / 敏感密钥被读),发现即告警,明确确认后按「一次一个」终止目标 agent;事件写入只追加 guard-events.jsonl,并通过 DSH host 工具 + MCP(stdio) 供其它 agent 查询
npm BSD-3-Clause -
Spending guard for AI agents that pay x402 paywalls in USDC: budgets, payee allowlists, price-spike checks and human approval before anything is signed. MCP server. Arc + NEAR testnet.
npm MIT -
An MCP guard where an AI agent's yes means exactly what the person saw. Reads pass; every other tool call waits for a coded yes.
MIT -
Guarded Kubernetes MCP gateway where AI agents observe, plan, and propose remediation that execute only after human approval and digest-bound verification.
NOASSERTION ★ 1 -
MCP server that gives AI agents governed, audited access to a legacy ERP's PostgreSQL database: read-only queries, guarded writes, atomic transactions and table snapshots.
MIT -
MCP proxy mode for Lodestar Guard — wrap any MCP-speaking agent (Claude Code, Cursor, Aider) so every tool call passes through the Action Kernel and every result through the Cognitive Core. Part of Lodestar, the trust layer for AI agents.
npm Apache-2.0 -
Enhanced distribution of HOL Guard: AI agent runtime security with MCP server, African threat intelligence, benchmarks, and architecture visualization
NOASSERTION -
x402 preflight CLI + the ZERO Payment Guard MCP server/x402 API source: check payment terms, Base USDC payments, tokens and wallets before an agent pays.
MIT -
Zero-dependency trustless governance for the SERVICE side. An MCP server or API re-verifies a calling agent's signed request against the agent's own published policy — so an agent that ignores its own guard still cannot make your service act.
npm MIT -
@upoch/guardactive
Upoch Guard: verifiable checkpoints, recovery and rollback for AI agents on Robinhood Chain. The upoch command, an MCP server for Claude Code, Cursor and Codex, and the engine behind the Upoch plugins for ElizaOS and LangChain.
npm MIT -
MCP server for the Guard ecosystem — config validation, docs search and live threat detection for fastapi-guard, guard-core and guard-agent. Answers from the versions installed in your project, not from a model's memory.
pypi MIT ★ 3 -
A deterministic guard for physical MCP tools: signed one-time passes, approvals by elicitation, and a proxy that protects any MCP server from prompt-injected agents.
MIT -
leakrank-guardgithub.com/ProtoTeemo/leakrank-guardupdated this year
Scans code changes for leaked secrets and insecure config, with actionable fixes for AI agents. MCP server, docs for the leakrank-guard npm package.
MIT -
Next-Generation Core Features for GenPark & OpenClaw: Agent Wallet Guard, Cross-Store Basket Optimizer, Product TruthGuard, and Circle Trend Radar.
MIT -
Hares (حارس) — the guard MCP server: kernel-capped, bwrap-sandboxed filesystem, shell, and cluster tools for LLM agents.
pypi Apache-2.0 -
Safety-first agentic Oracle DBA skills for Claude Code: 24 tier-gated executable domains (RMAN, Data Guard, RAC, perf, security…) fused with Oracle's official knowledge base, native SQLcl-MCP + a custom guarded host-ops MCP server.
NOASSERTION ★ 7 -
subway-contextactive
On-demand multi-window context transit and workspace guard for AI coding agents (Claude Code, OpenCode, Antigravity)
npm MIT -
On-demand multi-window context transit and workspace guard for AI coding agents (Claude Code, OpenCode, Antigravity)
npm MIT -
LocalHandsgithub.com/newcovid/LocalHandsupdated this year
A guarded local MCP server that lets a cloud AI agent operate one real machine - path whitelist, command guard, audit log. 222 tests.
MIT -
razorpay-hackathon-agentic-gatewaygithub.com/ankankisku-lab/razorpay-hackathon-agentic-gatewayactive
Zero-trust payment gateway for AI agents: Ed25519-signed AP2-style mandates, two-phase reserve/settle on Razorpay, layered prompt-injection guards (regex + Llama Prompt Guard 2), FAISS retrieval, MCP tools, and a signed hash-chain audit ledger
MIT -
My site, and Warden: an AI agent guarding an on-chain vault. Two LLM guards, code vetoes and contract caps, a soulbound trophy, a live feed and an MCP server. Astro, Cloudflare Workers, Solidity.
-
Local stdio MCP proxy that guards what your agent sends to yaptime
npm UNLICENSED -
Positifhttps://gitlab.com/getpositif/positifno update data
The agent switchboard: identity, policy, loop-guard, and an append-only audit ledger.
pypi -
safefs-mcpgithub.com/TEKERGUL/safefs-mcpupdated this year
A lightweight rollback guard for AI coding agents and MCP clients
MIT ★ 1 -
boundra-mcpactive
Boundra — policy-guarded payment approval gateway for AI agents (MCP server)
pypi Apache-2.0 -
MCP server and guarded agent loop for data-quality checks on synthetic claims tables
-
Stateful ride-support agent: versioned claims, evidence-backed RAG, durable memory and guarded MCP tools.
MIT ★ 40 -
nulmgithub.com/3keggo3-hash/nulmupdated this year
Local-first MCP agent quality and execution layer with guarded local tools
pypi MIT -
mcp-github-agentgithub.com/fmorgan-111/github-mcp-serverupdated this year
MCP server giving AI agents controlled GitHub access with policy guard and audit trail
pypi MIT -
secure-cluster-mcpgithub.com/FlorianSp2000/secure-cluster-mcpupdated this year
MCP Server allowing guarded interaction for coding agents with Slurm Cluster, intended for ML research
pypi MIT ★ 1 -
io.github.GravityWell8/reliableno update data
Resilient LLM/Agent calls via MCP: circuit breakers, timeout, jitter backoff, retry-storm guard.
pypi -
fi.zxoar/zxoargithub.com/lockwood-launch/lockwood-protocolno update data
Agent-first trading on the ZXOAR launchpad (Uniswap V4): scan, vet, guarded buy/sell, atomic launch
npm -
MCP-native toolkit for autonomous Casper agents with a Treasury Guard demo
MIT -
Progressive NTFS disk analysis and guarded cleanup for AI agents. Windows EXE, JSON CLI, and MCP.
MIT ★ 2