- Conduid
- Marketplace
- #ai-safety
MCP servers tagged ai-safety
464 live MCP servers tagged "ai-safety", ranked by trust score. Tags come from package metadata and repository topics, so this list covers servers across every category that work with ai-safety.
AISecuritySecondBrain
AI/LLM security knowledge base — OWASP LLM Top 10, MITRE ATLAS, prompt injection, agentic security. An Obsidian vault: 92 interlinked, fully-cited pages.
AgentInterdict
AgentInterdict — runtime enforcement for autonomous AI agents. Trust context. Verify authority. Interdict unsafe action.
Secchecker
Static security scanner and analyser for AI agents, MCP tools, and LLM applications - OWASP LLM Top 10, and infrastructure checks. Zero runtime dependencies.
Project Systemu
An open-source AI agent that builds the tools it's missing, mid-task — under your governance. Pull capabilities at runtime instead of pre-wiring them. Local-first, MIT.
Truthlens
AI hallucination detector with formally verified trust scoring for LLM outputs. No API keys needed.
Pherix
Transactional resource runtime for AI agents — undo the reversible, gate the irreversible, against your backend's own state.
Awesome AI Learning
A curated collection of high-quality learning resources for AI, covering Prompt Engineering, AI Agents, MCP, Skills, Claude Code, and AGI.
Likenessguard
LikenessGuard is an open-source reference project for pre-generation consent enforcement in AI image generation. V2 demonstrates the concept and AWS architecture. V3 is the formal…
Pipelock Verify Python
Python verifier for Pipelock action receipts (Ed25519-signed, chain-linked)
Palyan Agent Skills
Agent Skills for multi-agent AI governance
Cloudguard
Reliability and safety harness for LLM cloud agents. Tool-selection accuracy vs. tool count, blast-radius guardrails, and prompt-injection red-teaming against a real AWS mock. Eve…
Deepseek Forge
DeepSeek Harness 开发锻造工坊：审批守卫、开发 Skills、GitHub/浏览器能力与 Token Watch 消耗监督，装上就能干活。
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.
Dhms Engine
DHMS / AgentFuse: Execution assurance layer for AI agents — verify tool-call decisions, dispatch boundaries, and execution outcomes.
Hermes Loop Engineering
Hermes Agent starter kit for safe scheduled, stateful AI-agent loops
Agentnotary
Notarize, govern, and audit AI agents — cryptographic seal, runtime guard, EU AI Act docs, AI-BOM, adversarial fuzzer
Delegated Fidelity
The handshake for agents acting on the open internet. Six layers say the authorization is legal; the seventh asks the only question that matters once it is — is the agent still yo…
Vector Plugin
Red-team scanning for LLM agents, wired into your AI editor — one plugin for Claude Code, Cursor, Codex, and Gemini CLI
Aisafepy
Capability-based IFC, streaming-native cascaded guardrails, and an eval-to-guardrail compiler for LLM agents.
Agent Compliance Scanner
Free GitHub Action: scan AI agent code on every PR for regulatory compliance against the Bidda registry
Lich
Ruthless, algorithm-backed code review. Tests AI-generated code in an isolated sandbox to mathematically prove it works before merging.
Switchyard
Detect when Claude silently falls back from Fable 5 to Opus 4.8, log it to a local tamper-evident ledger, and keep your work on the frontier model.
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…
Lucinlabs
Static analysis for AI agent tools. Reads the function body, not the tool description.
LLM Security Payloads
A defender's field guide to attacks on LLM apps — prompt injection, jailbreaks, system-prompt extraction, data exfiltration — mapped to OWASP LLM Top 10, each with its defense. EN…
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。
Awesome AI Agent Security
A curated, opinionated list of resources for securing AI agents, LLMs, and the Model Context Protocol (MCP) — attacks, defenses, tools, labs, and research. All links verified.
Fde101
A practical, book-style field guide for Forward Deployed AI Engineers covering LLMs, production RAG, agents, LangGraph, evaluation, LLMOps, security, code examples, and interview…
Agent Eval Service
Scenario-driven framework for evaluating demographic bias in single- and multi-agent LLM workflows. Tracks bias across sequential stages, separates model non-determinism from syst…
Tool Output Spoofing Lab
Benchmarking schema-valid false tool observations and defense baselines for tool-using LLM agents.
Rtrace Agent Reliability
R-TRACE: Stateful agent evaluation and release reliability with reproducible Windows GPU evidence.
Rzp Guard
A deterministic authorization boundary that stops an AI agent issuing refunds nobody approved. Sits in front of Razorpay's MCP server; measured recall and false-positive cost publ…
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.
Agent Governance Toolkit
Manage and monitor AI agents in production environments with tools for governance, safety, and compliance.
Consentinel
Static security auditor for MCP servers. Finds tools that can do irreversible or external damage and verifies each one actually enforces a human-in-the-loop safeguard — not just d…
Pontifex
Build Enterprise-grade MCP servers
Observability Connector Platform MVP
A no-code platform for building, managing, securing, and observing Model Context Protocol (MCP) systems.
Beholder
The foundational SDK for algorithmic AI orchestration. The central nervous system powering the Enchanter ecosystem with native MCP support.
3 Distributed System MCP AI Incident Ethics Governance
🕸️ 3- Distributed MCP-based platform for AI incident analysis, with structured communication and traceability to support AI governance, Responsible AI and ethics-by-design.
Deskclaw
An AI customer-service agent that runs a real (mock) shop over WhatsApp — with hard safety rails. Skills → typed MCP tools → shared data, on OpenClaw.
Policy Guard
Policy and safety middleware for MCP servers — authorization, rate limits, confirmation gates, audit logging, and PII redaction.
Aip Python
Agent Identity Protocol Python Implementation
Agent Monitor
Local-first activity monitor, deterministic policy gate, and universal security boundary for autonomous AI agents and Model Context Protocol (MCP) clients.
Hmpo MCP Server Practice
HMPO application-status MCP server practice — FastMCP + SQLite. Parameterised read-only tools, guidance resource, JSON-lines audit log. MIT licensed. Topics: mcp, fastmcp, model-c…
Perimeter
Security and cost audit for your AI agent toolchain: MCP servers, hooks, skills
Pipsync MCP Evals
No-execution MCP mock and safety evaluations for trading-oriented tools.
FinLM
A governance layer for AI agents. Decides what the agent may do and may claim; it can't overrule either. Wired to a live broker. Risk gate I broke 4 times myself.
Aobench
Role-aware, permission-enforced benchmark for LLM agents that operate HPC systems — SLURM, telemetry, RBAC, docs, facility. A policy violation hard-fails the task. 88 tasks (10 ca…