- Conduid
- Marketplace
- #agent-security
MCP servers tagged agent-security
378 live MCP servers tagged "agent-security", ranked by trust score. Tags come from package metadata and repository topics, so this list covers servers across every category that work with agent-security.
Cross Provider Egress Guard
Destination-aware, default-deny egress control for AI coding agents, at the tool-dispatch layer across Claude Code and Codex from one shared policy.
Pfc MCP Authority Gap
Execution-Time Authority Integrity (ETAI) for MCP: a runnable demonstration that an agent can act on revoked authority under stateless OAuth, and a verifiable-authority layer whos…
Are Agent Integrations
Govern MCP and agent framework tool calls with ARE Foundation passport, scope, and policy checks.
Cinch AI Security
MCP server + cross-harness skills for building and operating AI agents safely. Grounded in NIST AI RMF, CISA, OWASP, CUSTODY, LASM.
Toolbastion
A zero-trust security gateway that inspects, approves, and audits MCP tool calls before execution.
Cupel
Assay every skill, plugin, and MCP server your coding agent loads. One local command grades safety, token cost, and quality.
Bluerock Security
Bluerock Security — independent third-party profile of a public API surface, by API Evangelist. BlueRock Security is a runtime security, observability, and governance platform for…
Ardra
Ardra is an all-in-one External Attack Surface Management (EASM) tool that helps identify and assess security risks by automating subdomain discovery, vulnerability scanning, DNS…
AgentInterdict
AgentInterdict — runtime enforcement for autonomous AI agents. Trust context. Verify authority. Interdict unsafe action.
Polaxis SDK MCP
Python SDK and MCP server for Polaxis - AI agent governance platform. Evaluate every tool call against policies in real time: block dangerous actions, enforce budgets, route to hu…
Ramen AI Integrations
Pre-execution semantic firewall for AI agents. Signed policy verdicts, BYOK, zero-trust receipts.
Capagate
Capability-aware firewall and policy compiler for MCP tool calls.
A Map
Cryptographic mandate chains and request signing for AI agent-to-agent trust. No central server. No shared secrets.
7h3 Protocol
Per-message signing + replay protection for MCP and A2A traffic. TypeScript / Python / Rust. Wire version aip/0.1.
Pipelock Verify Python
Python verifier for Pipelock action receipts (Ed25519-signed, chain-linked)
Velvet Rope
Test whether blocking one AI-agent route actually prevents the real-world effect. ShadowPath maps and replays alternate authorization paths.
Agent Control Plane
Read-only discovery and evidence-backed inventory for AI agents, tools, identities and MCP servers.
Daemonaudit
Red/blue security audit for self-hosted AI agent daemons — finds the secrets, the open doors, and the blast radius.
Sinon
AI agent pentest kit: a methodology and 54-probe corpus for prompt injection, tool abuse and over-permissioned agent actions. Verdicts rest on observed facts, not a judge model.
AgentCensus
Read-only scanner that inventories AI agents, tools, and credentials already running on platforms you operate (ServiceNow today), maps their dependencies, and flags what's ungover…
Dhms Engine
DHMS / AgentFuse: Execution assurance layer for AI agents — verify tool-call decisions, dispatch boundaries, and execution outcomes.
Batesian
Active security scanner for A2A and MCP servers
Aisafepy
Capability-based IFC, streaming-native cascaded guardrails, and an eval-to-guardrail compiler for LLM agents.
Command Scope Contract
Bounded shell and CLI execution for AI agents: structured contracts, policy-gated execution, hardened Linux runtime enforcement, and signed receipts.
Promptaudit
promptaudit
Attestload
attestload
Autonomyproof CLI
Prove what authority your AI agent holds — and fail the PR that grants new unsafe authority. Open-source, local, deterministic.
AgentTrustOps
Side-effect governance, approvals, crash-safe idempotency, reconciliation, and audit for AI agents.
Skillsentry
A semantic scanner for malicious MCP servers and agent skills — prompt-injection, tool-poisoning, rug-pulls, excessive agency.
Trifecta
Find and block the lethal trifecta that forms across your MCP servers.
Agent Webhook Check
Free single-file auditor for webhook signature verifiers — the highest-frequency real bug in AI-agent tool callbacks. Stdlib only, no deps. MIT.
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.
Locale
Agent policy layer for safe tool use across MCP, A2A, and orchestrated agents.
Skill Bouncer
TSA for agent skills: scan SKILL.md files before install, deny overreach with readable receipts, and bundle the safe ones.
Tool Output Spoofing Lab
Benchmarking schema-valid false tool observations and defense baselines for tool-using LLM agents.
Agent Permission Diff Bot
Detect and explain agent-facing permission changes in pull requests.
Dvah
DVAH — Damn Vulnerable Agent Harness: a patch-the-runtime security lab for AI-agent platforms. Exploit a real architectural bug, trace it, patch the harness, and prove the securit…
Handoffprobe
Open-source security testing for A2A → MCP agent handoffs and cross-protocol AI agent systems.
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…
Agentward
Audit what your AI agents can actually do to your machine. Scans installed MCP servers, skills, hooks and agent configs for leaked credentials, hidden prompt injection, exfiltrati…
SkillSpectorWeb
Skill Spector Web: A Web Assemby port of https://github.com/nvidia/skillspector
Guardrail Core
Policy enforcement for AI agent tool-calls: spend caps, rate limits, recipient allowlists, PII/secret detection, append-only audit log.
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…
Bureau
A hierarchy of AI agents that take real actions — shell, GitHub, email, purchases — behind an approval floor written in code, not configuration (Latch). Zero dependencies.
Agentgate MCP Firewall
Governance proxy for MCP tool calls with narrow independent verification for governed write_file and delete_file effects.
Agent Threat Scanner
面向 AI 智能体生态的安全扫描器，支持检测 Skills、MCP 服务、智能体配置及工具集成中的安全威胁。/ Privacy-first security scanner for AI agent artifacts, including Skills, MCP servers, agent configurations, and tool integr…
Kotro Proxy Engine
Local coding-agent control plane — MCP action governance + LLM-path protection + cost control. One Rust binary, no SaaS.
Securemcp Lite
A lightweight local MCP firewall for AI agents that enforces YAML security policies on tool calls.
Telekey MCP
Capability passports for MCP agents — delegated authority that can only narrow, never widen. A 2026 revival of General Magic's 1994 Telescript permit model.
Ganimedes Project
A local, zero-dependency security gateway for MCP servers. Audit, block and approve what your AI agents actually do.