- 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.
Jak Shield
Secure AI agent tool calls by blocking destructive actions, redacting sensitive data, and enforcing human oversight with this universal security gateway.
Soc Triage Agent
Local defensive AI agent that triages synthetic SOC identity alerts (LangGraph + MCP), with a governance-as-code layer - security manifest, control matrix (OWASP LLM Top 10 / MITR…
SafeDataBaseMCP
An MCP server with guardrails on real database infrastructure - SQLite, Postgres, and AWS RDS: every write is gated behind a propose/confirm preview the caller cannot skip, and RD…
AI Development
Enterprise AI-guardrails template. Shared CLAUDE.md/.cursorrules policy, secret-hygiene scaffolding, and MCP deployment checklists for human and AI coding agents.
Aeg Intent Gate
A tiny TypeScript approval gate for AI tool calls, MCP tools, and human-in-the-loop workflows.
Sqlgate
A deterministic, fail-closed write gate for LLM-generated SQL, exposed over MCP. The model can't talk its way past it.
Trusted Transcription
Catches an AI's confident mistakes before they reach a client — what turns an impressive demo into something you can invoice.
AI Policy 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
Sqlproctor
Deterministic semantic verification for agent-generated SQL.
Jevguard MCP
Official Model Context Protocol (MCP) server for JevGuard and TypeSafe AI
Retirement Answer Check
Checks AI-drafted answers to US retirement-account questions (SEND or REVIEW, with an IRS or FINRA source for every flag), plus a 2026 contribution-room calculator and MCP tool.
SentryQuery AI
Hallucination guardrail for AI output about indexed documents. POST a claim, get PASS or FAIL with the exact source passage, checked by a Researcher + Critic LangGraph pipeline (L…
Crewai Agentlock
Per-tool authorization for CrewAI agents. Every tool call gated, logged, and cryptographically signed.
Phi Guard MCP
MCP server and CLI for detecting, redacting, and auditing PHI before medical text is sent to AI agents.
Mainsheet
An agents runtime. Define an agent in one file, run it governed, prove what it did. A policy gate decides before every tool call in code the model never sees; budgets cap cost and…
Codapult Guard
Local-first architecture guardrails for JavaScript and TypeScript projects, built for AI coding agents and Next.js SaaS.
Blastdoor
Blast radius control for autonomous agents: capability boundaries, dry-run by default, verifiable kill switches, and an audit trail for every mutation.
Agentic Patterns
Reference implementations of 12 agentic AI patterns (ReAct, RAG, MCP, guardrails, multi-agent, and more) from canonical form to 2025-2026 research. Runs offline with no API key.
Clinical Agent Eval Demo
A deployment layer for a guardrailed clinical conversational agent: real FHIR over MCP, PHI redaction, a hash-chained audit trail, and an evaluation harness measured on held-out d…
Approvals
MCP server implementing a human-in-the-loop approval gate for AI agents, with a tamper-evident hash-chained audit log.
Guarded Tools
MCP server treating a large tool surface as a governance problem: tool-search discovery, request and token budgets, human-in-the-loop confirmation for writes, and an append-only a…
Multi Agent Market Research
Multi-agent LLM research platform: parallel data agents, synthesis agents, deterministic guardrails, MCP server
LLM Nexus
A transparent HTTPS proxy between any AI client and upstream LLMs — redacts PII, compresses prompts, caches semantically similar responses, and surfaces everything in a real-time…
Bio Harness
I want to explore how NVIDIA-labs Object Oriented Agents written in python compare with a LangGraph approach for bioinformatic pipelines and the deterministic execution layer for…
Hermes Plugin Senzu
Hermes Agent plugin: hands a stuck assistant over to the people who maintain it, and gates critical actions.
Frontend AI Patterns
Angular and TypeScript patterns for AI frontend systems: streaming UX, RAG citations, tool calling, MCP tools, approvals, state machines, and guardrails.
Nuvrail
Approval gateway for AI agents' email: an IMAP/SMTP proxy that stages every write (move, delete, send) for human sign-off. Self-hosted, AGPL-3.0.
Simdiff
The effect layer for AI-agent firewalls — simulate an action, get a canonical effect delta. Zero-dep, deterministic, fail-closed. 5 domains (fs, sql, shell, http, solana).
Dvara Examples
Reference Docker Compose stacks, cloud deploy recipes, and SDK integration samples for the DVARA LLM Gateway
Aitdp
AI Threat Detection Protocol — open spec, community ruleset and reference SDK for detecting prompt injection, jailbreaks, data exfiltration and tool abuse in LLM apps and AI agent…
Recruiting Agent
Guarded multi-turn recruiting agent: MCP server, LangChain MCP adapters, LangGraph, conversational evals with guardrail ablation, FastAPI, Docker, CI.
Agentsafe Guard
Runtime governance for AI agents. Wrap a tool in three lines; every call is signed and decided outside your process. Blocked calls never reach your code.
Nightcap
🌙 A hard stop for agents you leave running: dollar caps, pinned models and a crash-safe action ledger for MCP tools and Claude Code
Awesome Agent Safety
Awesome Agent Safety: curated agent-specific safety - threat models, guardrails, MCP/A2A security, evals, incidents, governance
Future AWS Agent Engineer
Multi-agent customer support on AWS: a Strands + Bedrock AgentCore agent team (inventory, refund, RAG policy, comms) behind a Cognito-secured chat app - with Bedrock guardrails, a…
Remedy Agent
An enterprise-grade, autonomous AI agent built to detect, audit, and remediate cloud security misconfigurations (like public S3 buckets) in a completely isolated, simulated AWS en…
Groundlens Dev
Guardrails for LLM answers. GroundLens checks an LLM answer against its sources without a model, decides PASS, REVIEW or FAIL under your policy, and keeps a signed record.
Jevmem
jevmem saves the decisions, rules and failed approaches from your Claude Code chats to JEVMEM.md in your repo, and brings the relevant ones back next session.
IDE Scanner Web
GuardRails by Abscissa: VS Code extension-package security, exact-release evidence, workspace monitoring, and a read-only OAuth MCP server.
Askdata Semantic Agent
Text-to-SQL analytics agent: semantic layer, 3-layer SQL guardrails, eval harness (47 golden Qs) and MCP server. Runs offline; optional OpenAI/Ollama planner.
Agent Workflow Training
Installable Claude Code guardrails + ai-workflow skill — drops deny rules, PreToolUse hooks, and a referenced iteration-loop skill into ~/.claude and project .claude. Cross-platfo…
Agent Cordon
Zero-dependency prompt-injection and data-exfiltration firewall for LLM agents: scans untrusted text and outbound actions, sees through obfuscation, guards the MCP/tool boundary.
TripMate AI
✈️ TripMate AI: multi-agent LangGraph travel planner in 3 versions, one per branch. v1 hand-written tools, v2 MCP servers, v3 supervisor routing, input guardrail and human-in-the-…
AI Projects Inventory Copilot
Governed GenAI inventory agent: provider-agnostic (LM Studio, OmniRoute, Amazon Bedrock), guarded text-to-SQL, RAG, MCP, guardrails, human-in-the-loop orders, rigorous evals with…
Agentx Security SDK
Runtime action firewall for AI agents: the keyless Shield. Screens catastrophic tool calls (DROP TABLE, SSRF, secret exfil) in-process, no key. Watches by default; set the posture…
Deflect
Customer support agent that answers, acts on or escalates tickets, with a 120 ticket eval suite, guardrails in Python before every tool call, human approval for large refunds and…
Mayi
Argument-level authority for AI agent tool calls. Every argument has one owner (app, model, human or derived). Plain deterministic rules, no judging model. Held calls wait for a p…
Aidd
AI-Driven Development: you decide what to build and what ships, and an AI delivery team does the rest and proves it. Four decisions stay yours: the plan, Spec approval, the releas…
emektor/toolwrit
chained audit log with RFC 3161 timestamp anchors. Also usable as a TypeScript or Python library.
MCP
Free, open-source guardrails and verifiable evidence for AI coding agents: Claude Code, Cursor, Codex, MCP, and GitHub. Scopebond Cloud is a separate proprietary hosted workspace.