Hush Security
Hush Security is a platform for discovering and governing AI agents and other non-human identities. It maps agent identities, credentials, permissions, tools, and infrastructure connections, helping security and identity teams assess risk and apply policy controls. The product is intended for organizations deploying AI agents and provides agent inventory, ownership attribution, activity auditing, access governance, and policy enforcement. Hush provides controls for limiting agent permissions and reducing reliance on static credentials, with visibility across agent and developer environments including Claude Code, Cursor, Codex, and Microsoft Foundry. Teams can use the platform to identify agents and credentials, review access, define scoped permissions, and monitor agent activity. Its free tier covers up to 25 protected agents, 50 access policies, and 10,000 discovered identities; paid plans support broader usage.
Learn more about Hush Security
Harden
Harden AIF is an agent endpoint security platform for AI coding agents. It evaluates supported agent tool calls before execution using the developer’s intent, session context, organisational policy, and the effect of the proposed action.
Harden helps protect against destructive commands, unauthorized access, unintended data transfers, secret exposure, privilege misuse, and other unsafe agent actions. Legitimate actions can proceed normally, sensitive data in supported flows can be safely redacted, and actions that fall outside the developer’s intent or authority are blocked before execution.
Harden works across popular coding agents and agentic development tools including Claude Code, Codex, Cursor, Antigravity CLI, Kiro, Hermes, and OpenClaw, providing a consistent security layer across the agent ecosystem.
Learn more about Harden
Mindgard
Mindgard is an automated AI red teaming and offensive security platform purpose-built for AI systems and agents. Built for security and engineering teams accountable for AI in development and production, it continuously red teams models, agents, and applications to find the vulnerabilities most likely to cause a breach, validate whether guardrails and controls hold, and show teams how to close each gap.
AI red teaming tests emulate real attacker behavior across prompt injection, jailbreaks, data extraction, poisoning, and agentic tool abuse, and map findings to OWASP LLM Top 10, MITRE ATLAS, NIST AI RMF, and the EU AI Act. Unlike traditional AppSec tools or manual penetration testing, Mindgard cuts AI risk assessment from weeks to hours, pinpoints and validates defensive gaps, and supports continuous policy compliance as AI systems change. Capabilities: AI Discovery & Recon, AI Red Teaming, AI Assessment, AI Runtime Protection, Model Scanning, AI Governance & Compliance. SOC 2 Type
Learn more about Mindgard
AgentShield
AgentShield is a next-generation identity platform built to verify both human users and AI agents acting on their behalf. It enables organizations to confirm who an agent is, whether the person behind the agent has provided explicit authority, and that the agent is trustworthy, all through APIs and JavaScript integrations. The product includes tools that detect agentic sessions on a website. and enforces identity and permission checks for agent-to-agent or agent-to-service interactions under the open Model Context Protocol Identity (MCP-I) specification. With KYA, businesses can securely manage agent identities and permissions, institute audit-trails, automation workflows, and finely-tuned access control for autonomous systems, thereby protecting themselves from misuse of digital identities and ensuring transparency when AI systems act on behalf of users.
Learn more about AgentShield