According to a recent LinkedIn post from Noma Security, the company is spotlighting new research on security risks associated with AI agents, Model Context Protocol (MCP) servers, and agent skills. The post cites internal analysis of hundreds of MCP servers and skills, claiming that a majority exhibit high‑risk characteristics and potential for arbitrary code execution.
Claim 55% Off TipRanks
- Unlock hedge fund-level data and powerful investing tools for smarter, sharper decisions
- Discover top-performing stock ideas and upgrade to a portfolio of market leaders with Smart Investor Picks
The LinkedIn post highlights a research report titled “Lethal by Design,” which focuses on how design choices in extending AI agents shape their attack surface and the ability of security teams to monitor and intervene. It also references a “No Excessive CAP” framework that is presented as guidance for governing what agents can do, decide, and access in enterprise environments.
For investors, this emphasis on AI agent security research suggests Noma Security is positioning itself as an early specialist in a rapidly emerging risk category around “agentic AI.” If the underlying findings gain traction with security and AI engineering teams, the company could benefit from increased demand for tools and frameworks that address observability and control in complex AI deployments.
The post’s focus on empirical analysis and governance frameworks may help Noma Security differentiate from more generic cybersecurity vendors by anchoring its brand around research‑driven insight. This could support premium pricing, deepen engagement with large enterprises experimenting with AI agents, and potentially expand the company’s addressable market as regulatory and compliance scrutiny over AI operations grows.
However, the commercial impact will depend on how effectively Noma Security converts research visibility into product adoption and recurring revenue. Investors may want to watch for subsequent product releases, customer case studies, or partnerships that explicitly tie the “Lethal by Design” and No Excessive CAP concepts to deployable solutions and measurable risk reduction in production AI environments.