- 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.
Warden CLI
Developer CLI for Warden — init, start, audit, policy test, scan
Atbash
Atbash safety judge exposed as a standalone MCP server
Webcite MCP Server
WebCite MCP Server - verify AI claims against authoritative sources, bind quotes back to their source passage, and extract document figures deterministically. Works with any MCP-c…
Audit
A first-pass linter for MCP servers: spec conformance + safety heuristics, in one local zero-dependency CLI.
Taintgate
🚧 A policy gate for AI agent tool calls, with provenance tracking. Blocks what prompt-injection classifiers miss: where an instruction came from.
Sanna
Trust infrastructure for AI agents — constitution enforcement and cryptographic receipts. Python SDK.
Revl
A language for safe, universal spatiotemporal composability (Cordis paradigm) and orchestration.
Relayops
Production-shaped control plane for AI support agents: scoped routing, policy broker, tool boundaries, approval queue, replay verification, audit traces, and human handoff. Protot…
Visor
Runtime policy enforcement and audit control plane for MCP tool execution. Deterministic, non-AI policy engine that intercepts MCP tools/call requests before execution.
Narthex
Prompt-injection defenses for Claude Code. A PreToolUse Bash hook blocks compositional credential-exfiltration shapes (secret read plus network, env dump to network, remote script…
Agentseal MCP Intel
MCP server for AI security intelligence. Check any MCP server for supply-chain threats before installing -- from Claude, Cursor, or Windsurf.
Argos MCP Guardrail
A lightweight, sub-millisecond security shim that prevents AI agents from accessing sensitive files (.env, SSH keys) and executing destructive commands via MCP.
Qed Proof Core
Proof of Agent Work (PoAW): an open protocol and self-hostable node that verifies what AI agents claim they did and issues signed, publicly checkable receipts.
Sovereign MCP
Deterministic MCP Security Architecture. FrozenNamespace as Root of Trust for Model Context Protocol tool verification
Sovereign MCP
Deterministic MCP Security Architecture. FrozenNamespace as Root of Trust for Model Context Protocol tool verification
Agent Firewall
Lightweight, zero-dependency AI Agent runtime firewall (L7 HTTP Proxy + MCP Sandbox) in Go.
AI Projects
AI/ML engineering portfolio covering agentic AI, multi-agent orchestration, RAG, AI safety, knowledge graphs, document intelligence, voice AI and classical NLP/ML.
Sosa
SOSA™ — Supervised Orchestrated Secured Agents. A governance framework for production-grade autonomous AI operations.
Vanta
Open-source local AI operator: desktop app, terminal, messaging, and automation with a Rust safety kernel that gates every action.
AgentInjectionBench
A benchmark for evaluating prompt injection attacks in agentic tool-use pipelines
ROEGATE
The first reference monitor for autonomous AI penetration testing agents. Mandatory enforcement, not prompt-based guardrails. Patent pending.
Bounded Agent Runtime
Security runtime for AI coding agents: bounded authority, exact-candidate evidence, separate review and human gates. Works with Codex, Claude and OpenCode.
Freshctx
Pre-action evidence revalidation for AI agents. Detect stale reasoning before agents write, send, update, or delete.
Dynolab
Dyno Lab — an AI safety, alignment and interpretability research workbench for Apple Silicon. Explore activations, probes, SAEs and interventions with local inference, a Python SD…
Diplomat Gate
Runtime approval gates for AI agent tool calls. Intercept payments and emails before execution
AI Learning Workbench
面向本科生的低成本 AI 学习工作台：整合 Claude Code、Codex、MCP 与 Skills，提供实验报告、课程作业、Python 入门和安全配置教程。
Hermes Rubric
Evidence-first LLM-as-judge scoring for AI artifacts — papers, PRs, prompts, cold emails: synthesizes a rubric, collects quoted-evidence citations, scores only against that eviden…
Actiongate Jev
Runtime authorization and guardrails for AI-agent tool calls with deterministic policy and TypeSafe Jev via OpenRouter.
Agent Trace Witness
External witness for autonomous multi-agent AI systems: signed readiness seal, choke-point capture, and PROV-DM causal graphs for post-incident reconstruction
Jev Engineering
The decision layer for AI agents. Typed, calibrated decisions in ~400ms for two hundredths of a cent: gate tool calls, route models, rank options. With the 300-call injection test…
Kiri Gate
A gate in front of your AI agent's tools: acts on what can be undone, asks about what can't.
Common Sense Skill
One file. 9 questions. Stops AI agents from doing the obviously-wrong thing. MIT. Works with Claude Code, Codex, Cursor, OpenClaw.
Jev Compactor
Context compaction and safety gating for AI agents via TypeSafe Jev: keeps messages verbatim, no summarization. OpenAI, Anthropic, LangChain, CLI, MCP.
Provio
A safety floor your AI agents can't get under: one policy for Claude Code, Codex, Gemini CLI, Cursor, Windsurf and MCP, judged by what each call will run, backstopped by the kerne…
Guard Layer
Runtime guardrails between an AI agent and its tools: scan what it reads, gate what it does, and track the session so a secret read earlier can't be exfiltrated later. Containment…
Rustmachineguard
AI-agent posture scanner for developer machines: inventory, sourced threat catalog, live MCP probing, scan diffing, CycloneDX 2.0 Blueprint, compliance evidence. Local only. Not a…
Hermes Sentinel
Hermetically-sealed secret protection and capability execution MCP for AI tools
Enlil
Source-available control and audit plane for AI agent actions. Enforce what your agents may do, and prove what they did.
Goop Shield Community
Runtime defense for AI agents. 24 inline defenses, 3 output scanners, MCP server, framework adapters.
Groundcheck MCP
MCP connector that verifies a claim's grounding against reality (quotes, citations, code, repos, math) with no LLM in the verification path. Works in Claude and Gemini.
3 Surgeons
Multi-model consensus system — 3 LLMs cross-examine each other to catch blind spots on critical decisions
LLM Provost
Compliance and governance proxy for LLMs and MCP servers. Every chat must register. Every action is logged. Lua Rules are applied. From anarchy to audit!
Sentinel
OSS safety runtime for sponsored recommendations in AI conversations; API/MCP verification, signed receipts, deterministic gates.
Lore
Project-scoped rule memory for coding agents.
Aegis Guardrails
Deterministic governance and guardrails for autonomous AI coding agents. Features AST verification, evidence-based rule execution, and native MCP support for Antigravity and Claud…
Shield
Security audit framework for MCP servers. 17 detectors, confidence scoring, 3 scan modes, zero dependencies. pip install mcp-shield-audit
Fuse
A local firewall for AI agent commands
Headroom
More human oversight can make an AI agent less safe. Headroom is a human-in-the-loop firewall for coding agents that measures when to trust the human: oversight as resource alloca…
Pipelock Verify Python
Python verifier for Pipelock action receipts (Ed25519-signed, chain-linked)
Hydra
Real-time AI threat interception. Blocks CVE-mapped attacks, poisoned configs, and destructive terminal commands before execution.