CIS Introduces Guidance for Securing AI Agents and Autonomous Systems
Summary
CIS, Astrix Security, and Cequence Security developed three AI security companion guides. The AI Agent guide addresses risks from tool use, memory and retrieval, multi-agent workflows, and runtime ownership.
Key points
- The AI Agent Companion Guide extends CIS Controls guidance to AI systems that take actions through tools, APIs, and automated workflows.
- Recommended safeguards include strict tool allowlisting, scoped capabilities, access controls for tool calls, human approval for high-impact actions, and audit trails.
- The guide treats agent memory and retrieval systems as security-sensitive because poisoned or manipulated data can influence decisions.
- It highlights multi-agent risks including cross-agent contamination, unclear authority boundaries, misplaced trust, and runaway loops.
- Runtime ownership—provider-managed, enterprise-managed, or local—affects control over agent memory, logs, tools, and behavior.
Article Details
- Event Type
- Release of the AI Agent Companion Guide, a cybersecurity guidance publication for securing AI agent environments.
- Impact
- The guide provides organizations with recommendations for controlling agent actions and tool access, protecting memory and retrieval systems, managing multi-agent risks, and securing runtime deployments. The article describes potential risks, not a reported incident or confirmed compromise.
Vendors
Astrix SecurityTo address these challenges, the Center for Internet Security® (CIS®), Astrix Security, and Cequence Security partnered to develop actionable cybersecurity guidance tailored to AI environments.Center for Internet SecurityTo address these challenges, the Center for Internet Security® (CIS®), Astrix Security, and Cequence Security partnered to develop actionable cybersecurity guidance tailored to AI environments.Cequence SecurityTo address these challenges, the Center for Internet Security® (CIS®), Astrix Security, and Cequence Security partnered to develop actionable cybersecurity guidance tailored to AI environments.
Products
AI Agent Companion GuideThe result is three new CIS Companion Guides: the AI Large Language Model (LLM) Companion Guide, the AI Agent Companion Guide, and the Model Context Protocol (MCP) Companion Guide.AI Large Language Model (LLM) Companion GuideThe result is three new CIS Companion Guides: the AI Large Language Model (LLM) Companion Guide, the AI Agent Companion Guide, and the Model Context Protocol (MCP) Companion Guide.CIS Critical Security ControlsThe work extends the globally recognized CIS Critical Security Controls® (CIS Controls®) into environments where autonomous decision‑making, tool and API access, and automated threats introduce new risks.Model Context Protocol (MCP) Companion GuideThe result is three new CIS Companion Guides: the AI Large Language Model (LLM) Companion Guide, the AI Agent Companion Guide, and the Model Context Protocol (MCP) Companion Guide.