#agentsecurity

New reference implementation shipped today: #LangChain middleware that gates AI agents from connecting to unfamiliar MCP servers behind a mandatory security check. It is not an optional tool call the agent is free to skip. https://github.com/nzdsf2-gif/relayshield-langchain-gate #InfoSec #MCP #AgentSecurity #AIsecurity

New from RelayShield: four #AIsecurity checks for agents built with smolagents, published as an MCP server on @huggingface.

• MCP server reputation check before your agent connects
• Prompt-injection pattern detection on ingested content
• Tech-stack CVE monitoring (covers agent frameworks themselves — Langflow's CVE-2025-3248 was the initial-access vector in the first documented autonomous-agent ransomware op)
• Bulk identity-risk scoring

Self-serve, pay-per-call, no subscription.

https://huggingface.co/blog/relayshieldadmin/smolagents-agent-security-tools

#InfoSec #MCP #AgentSecurity #ThreatIntel

Shipped today: RelayShield's AI-agent security checks are live as an MCP server on Hugging Face including MCP server risk, prompt-injection breach detection, agent framework CVE monitoring, and bulk agent-identity risk scoring. Same "Agentic Attack Surface" thesis behind our AgentCore/x402 work: security built for how agents actually get compromised, not humans. Live now: https://huggingface.co/blog/relayshieldadmin/relayshield-agentic-attack-surface-mcp #AIAgents #MCP #AgentSecurity #CyberSecurity