- Conduid
- Marketplace
- #guardrails
MCP servers tagged guardrails
304 live MCP servers tagged "guardrails", ranked by trust score. Tags come from package metadata and repository topics, so this list covers servers across every category that work with guardrails.
Pherix
Transactional resource runtime for AI agents — undo the reversible, gate the irreversible, against your backend's own state.
Think Tank Skill
🧠 Think-tank is a host-enhanced workflow skill framework that can be adapted across projects.
Agentic Refund Guardrail
Safe refund policy layer for AI agents — Python + TypeScript. Same behavior, shared tests.
Clinical Doc Intelligence
AI-powered FDA drug label intelligence platform — production RAG with 5-stage retrieval, multi-agent orchestration, clinical guardrails, and 54 automated tests
Citadel Saas Factory
265 autonomous AI agents across 15 business domains. Full-stack SaaS framework with real model routing, RAG pipeline, guardrails, and Chrome extension. FastAPI + Next.js + K3s. Ru…
Tool Call Guard
Deny-by-default policy gate for AI agent tool calls — allowlists, validation, rate caps, approval hooks, audit trail. One policy model, enforced in JavaScript and Python.
AgentToolGate
本地 AI Agent 工具治理网关：不防提示词注入发生，防注入得逞后的高危工具调用后果（policy / approval / audit / MCP 双向治理）
Gate.cat
Block irreversible AI-agent actions before they run — deterministic, fail-closed action veto for Claude Code, Codex, Cursor, and other tool-using agents.
Agentnotary
Notarize, govern, and audit AI agents — cryptographic seal, runtime guard, EU AI Act docs, AI-BOM, adversarial fuzzer
Aisafepy
Capability-based IFC, streaming-native cascaded guardrails, and an eval-to-guardrail compiler for LLM agents.
Warrant MCP
Write the rules for your AI agent once, in plain English. Deterministic code checks every tool call, and a Claude Code hook blocks what your rules forbid. Ships a Claude Skill tha…
Zcode Autoresearch
ZCode 自主实验循环插件，让 coding agent 基于固定度量自动迭代优化
Prfrail
Provable rails for unattended AI engineering — every code change, gate, evidence and rollback is auditable. / ProofRail 为 AI 无人值守编程铺设可证明的轨道——代码变更、门禁、证据、回滚，全程可审计。
Munch Food Agent
A food-ordering agent that can't order food you didn't approve. Guardrails between the LLM and the world: hard dietary/allergen blocks, spend caps, and a mandatory human confirmat…
Guardrails
Enterprise-grade security gateway for Model Context Protocol (MCP). Multi-layered protection with DLP, UEBA, policy-based access control, and comprehensive audit logging for AI as…
Claude Code Operating Framework Example
Sanitized Claude Code operating framework teaser showing sub-agents, skills, hooks, MCP, guardrails, and evaluation.
Silmaril
Silmaril — independent third-party profile of a public API surface, by API Evangelist. Silmaril is a Y Combinator-backed runtime security company that builds an AI application fir…
House Rules
Standing rules for working with coding agents, plus the hooks and CI gates that enforce them. Vendored into every repo I work in.
Access AI Gateway
LLM gateway over a multi-tenant access-control system, with guardrails against indirect prompt injection and cross-tenant data exposure.
Redcell
Security layer for AI agents — score the system prompt, firewall untrusted input, screen every tool call. Deterministic, 0-API, runs at the edge.
Guardrail Core
Policy enforcement for AI agent tool-calls: spend caps, rate limits, recipient allowlists, PII/secret detection, append-only audit log.
Self Improving Agent Harness
An AI coding agent that fixes bugs in a sandboxed repo -- and a harness that learns from its own failures, writes guardrails against them, and proves with regression evidence they…
Yana AI
Yana AI 🐰 One runtime. Any AI. Human-governed. A local-first, cross-platform system for running, connecting, orchestrating, and governing AI — with deterministic control over wh…
Pymcp Test
AI-native commerce infrastructure that makes merchants discoverable and transactable to AI buyers, with bounded purchasing agents, intelligent cross-sell, agentic checkout with pa…
Drift Detector
Claude Code + Grok drift detector — scores turns vs your output contract and steers back.
Policy Guard
Policy and safety middleware for MCP servers — authorization, rate limits, confirmation gates, audit logging, and PII redaction.
Micro AI Projects
This repository is a collection of micro-projects, each designed to help me understand one infrastructure primitive at a time.
Guardrail Gateway
Security gateway for the Model Context Protocol: role-based access, prompt-injection and secret scanning, PII redaction, and a tamper-evident audit log in front of any MCP server.…
Dietary Exposure MCP
Auditable dietary exposure screening and evidence handoffs over MCP
Valta MCP
Financial guardrails for AI agents — pre-call spend gates and a cryptographically hash-chained audit trail. Stops an agent before it overspends, not after.
Promptwall
A firewall for LLM traffic: detect prompt injection, jailbreaks, and secret/PII leaks. Zero-dependency Python library + CLI + MCP server.
SQL Analytics
MCP stdio server for bounded read-only SQL with SQLite authorizer guardrails and protocol integration tests.
Agentx MCP
Keyless MCP security shield: wrap any MCP server so a poisoned tool call cannot wreck your system and your agent keeps working. Launcher for agentx-security-sdk.
Architect Guardrail
MCP-based AI governance layer for Cursor and Claude that injects company policies, approved libraries, and security rules into AI code generation workflows.
Cliguard
Experimental local-first guardrails for human and AI-generated CLI actions.
Approvals MCP
An MCP approval gate for AI agents
Provenant Examples
Runnable Apache-2.0 examples for Provenant — authorize, hold, or deny what your AI agents do, with a tamper-evident audit trail.
Idea Anvil
Best AI Coding Agent Pipeline Builder 2026 – Plan-Forge Drift-Proof Execution Contracts
Orgai Platform
Self-hosted AI compliance enforcement — one policy source enforced across Claude Code, Cursor, Copilot & any MCP agent, with git-hook + CI gates and an append-only audit trail
Ottrix
Lightweight TypeScript framework for LLM agents: ReAct loop, tool calling, MCP, multi-agent workflows, memory, guardrails & observability. Claude, OpenAI & Ollama via fetch — no v…
Supaproxy Core
AI operations platform. Route any LLM to any team through one governed layer.
Comfy Preflight
A gate that runs on a ComfyUI workflow graph in the seconds before submission and halts a submission that would spend credits producing a known-wrong result.
Apollo AntiPoison AntiHallucination Research MCP
A zero-cost FastMCP research server for AI agents featuring Dual-Pillar defense: 4-pass anti-poisoning / prompt-injection sanitization & anti-hallucination context budgeting over…
Chaperone
Open-source AI agent security gate: deterministic ALLOW/BLOCK/ESCALATE authorization before every tool call, plain-English policy compilation, tamper-evident audit ledger. MCP sec…
Synartesis
An undo layer for AI agents. Records what an agent changed through MCP, puts it back, and refuses to let it do the unforgivable unasked.
Prompt Injection In LLMs
Implemented and evaluated protection mechanisms to determine their effectiveness against direct prompt injection attacks.
AgenticAI Learnings
My hands-on learning journey in Agentic AI covering LangChain, LangGraph, RAG, MCP, Deep Agents, Guardrails, LLM Evaluation, and LLM Gateways.
Mcpguard Go
mcpguard-go: local AI agent security scanner (Go, zero-dep single binary)
Nekoi MCP
Judges Claude Code by its own transcript, and stops the attempt at the point of intent - not after the fact. A hook and MCP server in one Go binary.
Officer
Pit Officer - open-source pre-trade risk control plane. Checks orders against your limits and signs the approval the trade needs; never routes or fills an order. REST API, MCP ser…