- 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.
AgentCloak
Secure email proxy for AI agents — content filtering, PII redaction, and prompt injection detection over MCP
Orca AI Incident Archive
An open database of real-world AI agent incidents.
Memnox
Your agents can already do the work. Memnox makes it safe to let them do it without you. See what Claude Code, Codex and Cursor can actually do on your machine, and put the danger…
LLM Shield Proxy
Streaming privacy for LLM apps: pii-leak-benchmark (does your gateway leak PII in streams?), chunk-invariance (one-line test for split-boundary leaks), mcp-ssrf-check, and a self-…
Canary
Which fields of an MCP tool result actually reach the model's context, measured with benign canaries across every installed server
webcite-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…
Avakill
🔪 Open-source safety firewall for AI agents. Intercepts tool calls before they execute, enforces YAML policies, and kills dangerous operations in real-time. Works with OpenAI, Ant…
Hardened Skills
200 AI agent skills, hardened with targeted behavioral guardrails. Free drop-in replacements.
IgorGanapolsky/ThumbGate
action gates via PreToolUse hooks. Install: `npx thumbgate`.
MCTS
MCTS (Model Context Threat Scanner) is a local-first security scanner for MCP servers -- static and live tool discovery, multiple analyzers, auditable risk scores, and JSON, SARIF…
LLM Differential Privacy Gateway
Noisegate: a differential privacy gateway that lets an untrusted LLM agent query sensitive data over MCP (Model Context Protocol), with a formal guarantee no individual's record c…
Torque
Keep Salesforce consulting work easy to resume and hand over: a private workspace per client and initiative for your coding assistant, with delivery workflows, change records and…
Sentinel AI
Real-time AI safety guardrails for LLM apps. 10 scanners: prompt injection, PII, harmful content, code vulnerabilities, obfuscation detection. Sub-ms latency. Python + TypeScript…
Tardygrada
A verified runtime for trustworthy AI agents: every claim passes an 8-layer verification pipeline before it counts. Deterministic checks, explainable verdicts, no LLM judging an L…
Awesome AI Red Teaming Jp
AI Red Teaming / AI Safety に関する日本語リソースのキュレーションリスト
Sofagent
FDE Harness: audit-first governance for AI coding agents — 24 rules, HMAC chain, snapshot rollback (83 tools, 13 plugins)
Norviq
Runtime security platform for LLM agent tool calls on Kubernetes
Auditable
Audit any agent decision across its past, present, and future, on one typed graph.
SaneProcess
Workflow guardrails for coding agents and LLM-assisted development: AGENTS.md, native hooks, MCP, SaneMaster, circuit breakers, and shared process checks.
Artguard
Open-source AI artifact scanner. Detect malicious agent skills, MCP servers, and IDE rule files before they run.
Concealer
🔐 The local-only secret manager for the AI-coding era with MCP, CLI, Web UI, TUI and Extension. No cloud, No telemetry, No account.
Brutalist MCP
All AIs are sycophants.
Agent Audit Kit
Static scanner for MCP-connected AI agent pipelines. 332 rules across 14 categories, 12 compliance frameworks, OWASP Agentic 10/10 + MCP 10/10, GitHub Action, SARIF, public CVE-to…
Apoa
Agentic Power of Attorney (APOA) | Authorization infrastructure for AI agents
Lynx
Policy-gated, durable, audited execution for AI agent tool calls. Every action gets a five-verdict policy check, a crash-safe checkpoint, and a tamper-evident audit event — a libr…
Isnad
Grade every agent, scraper and model in a claim's chain — provenance, trust scoring and audit evidence for LLM pipelines
Erisml Lib
ErisML is a modeling language for governed, foundation-model-enabled agents operating in pervasive computing environments
VibeD
Deploy AI-generated web apps to your own Kubernetes cluster in seconds, over MCP. No container build on the deploy path.
Trustmodel
Score any AI for trust — Eval, Monitor, Govern. 10 trust dimensions, one free API key (5 credits / $500).
Prompt Armor
Open-source prompt injection detector — 5 layers, 91.7% F1, ~27ms, offline, Apache 2.0
Claude Guardrails
Hardened security configuration for Claude Code; permission deny rules, shell hooks, and prompt injection defense in full and lite variants.
Baseline
A discipline layer for Claude Code. Hooks at every tool boundary, a workflow that runs from intake to commit, and a small constitution the agent cannot bypass.
AI Safety Formalization Atlas
The open workbench for AI safety, made formal. Turn safety questions into machine-checked Lean proofs — a shared launchpad where researchers and AI agents build provable safety to…
Bonklm
MCP SDK connector for BonkLM
Guardian
MCP security scanner - detect prompt injection in tool descriptions
mcpwall
Deterministic security proxy for MCP tool calls. Blocks dangerous requests, redacts secrets from responses, catches prompt injection. No AI, no cloud, pure rules.
Mcpwall
Deterministic security proxy for MCP tool calls. Blocks dangerous requests, redacts secrets from responses, catches prompt injection. No AI, no cloud, pure rules.
bouncer
A deterministic exfiltration firewall for MCP tool calls — gates the destination, not the tool. No LLM in the decision path.
Hermes Prime
Drift-prevention session-init convention card for fresh Claude Code sessions. Injects a self-contained card so a new session opens with its grounding triggers, calibration rules,…
Proxy Guard
Sub-35µs security proxy and credential scrubber for Model Context Protocol (MCP) servers in Claude Desktop, Cursor, and Windsurf.
Limetry
MCP server for Limetry agent action governance and telemetry
Phantom Secrets
MCP server for Phantom Secrets — lets AI coding tools manage API keys safely without ever seeing real values. 10 tools for Claude Code, Cursor, Windsurf, and Codex.
Quality Gate
Quality gate for MCP servers — compliance, security, and efficiency testing
Reagent
Governance layer for Claude Code — policy enforcement, hook-based safety gates, and audit logging for AI-assisted projects
Rea
Agentic governance layer for Claude Code — policy enforcement, hook-based safety gates, audit logging, and Codex-integrated adversarial review for AI-assisted projects
Altais
Modular, open-source MCP server providing comprehensive security analysis for AI coding agents.
Free AI Ops
Free MCP server for developers adding human review gates, prompt risk checks, FAQ review, and CRM note safety to AI tools.
Prism Verify
prism-verify — runtime LLM verifier (family-different, reasoning-stripped, multi-lens, signed Ed25519 receipts). Zero-prerequisite npx install via a verified binary launcher.
Metago Lifeform
MetaGO Agent Harness（智能体运行时控制层套件 · 驭智层）— 智能体运行时控制层，让智能体从工具升级为守规矩、会进化、可追溯、能闭环的生命体。39 技能 · 53 MCP tools · 8 公理 · Engine V2（KMWI 四层记忆 + 元进化五阶段 + 技能智能生成）· 决策锁四道关卡 · 全链路溯源。支持 7 大 AI 编程…
Border Patrol
A policy-enforcing MCP gateway: one MCP server that proxies all your others and blocks dangerous tool calls