Key Takeaways
- • Ultra Security has emerged from stealth with $10 million in seed funding to build an agentic security platform that provides real-time visibility and guardrails for autonomous AI agents, according to a Business Wire announcement dated October 21, 2026.
- • The platform sits between AI agents and the MCP servers and connectors they call, inspecting every tool call, data read, and prompt request as it happens rather than after the fact.
- • Investors include Ludlow Ventures, Hidden Capital, Entrée Capital, Detroit Venture Partners, firedrop, and FDVC, along with angels from the security and SaaS communities.
- • The company was founded in 2025 by Chase Lee, former founder and CEO of Trustpage, which was acquired by Vanta, and its founding team comes from Vanta and Secureframe.
- • Ultra reports SOC 2 Type II, ISO 27001, ISO 42001 and AARM Extended conformance, and has completed a FedRAMP 20x Class B audit with a 3PAO while pursuing FedRAMP certification.
Summary
Ultra Security, headquartered in Birmingham, Michigan, has come out of stealth with $10 million in seed funding to address security gaps in autonomous AI agents, according to a Business Wire release. The company describes its product as an agentic security platform built for real-time visibility and guardrails, allowing organizations to see and control what agents are doing as it happens. Ultra's investors include Ludlow Ventures, Hidden Capital, Entrée Capital, Detroit Venture Partners, firedrop, and FDVC, along with angels from the broader security and SaaS communities.
The company positions itself between AI agents and the Model Context Protocol (MCP) servers and connectors they call. Every tool call, data read and prompt request passes through Ultra. The release argues that MCP is becoming the standard way agents connect to tools, APIs and sensitive data, and that adoption is outpacing the tooling built to monitor it. It contends that web application firewalls, API gateways and cloud posture tools were not designed to observe what an agent does inside a live session.
The platform is organized around four functions: guardrails that enforce plain-language policies on each tool call, including blocking secret and credential access, redacting PII, protecting PHI, rate-limiting abuse and preventing data transfer between connectors; governance over which agents, connectors and tools each person or team may use; observability that logs, traces and makes queryable every agent-to-tool interaction, with AI-native anomaly detection; and connectivity through a vendor- and model-agnostic bridge. Ultra states it runs locally by default, with tool calls, payloads and audit trails kept on the user's machine, and offers a remote-hosted option.
Founded in 2025 by Chase Lee, former founder and CEO of Trustpage (acquired by Vanta), the company says its team has spent the past year in stealth working with customers and design partners. The release claims Ultra is SOC 2 Type II, ISO 27001, ISO 42001 and AARM Extended conformant, aligns with NIST AI RMF and the Agentic Trust Controls, and has completed a FedRAMP 20x Class B audit with a 3PAO, with FedRAMP certification in progress. Access can be requested at ultra.security. These claims are the company's own and have not been independently verified in the source material.
The company positions itself between AI agents and the Model Context Protocol (MCP) servers and connectors they call. Every tool call, data read and prompt request passes through Ultra. The release argues that MCP is becoming the standard way agents connect to tools, APIs and sensitive data, and that adoption is outpacing the tooling built to monitor it. It contends that web application firewalls, API gateways and cloud posture tools were not designed to observe what an agent does inside a live session.
The platform is organized around four functions: guardrails that enforce plain-language policies on each tool call, including blocking secret and credential access, redacting PII, protecting PHI, rate-limiting abuse and preventing data transfer between connectors; governance over which agents, connectors and tools each person or team may use; observability that logs, traces and makes queryable every agent-to-tool interaction, with AI-native anomaly detection; and connectivity through a vendor- and model-agnostic bridge. Ultra states it runs locally by default, with tool calls, payloads and audit trails kept on the user's machine, and offers a remote-hosted option.
Founded in 2025 by Chase Lee, former founder and CEO of Trustpage (acquired by Vanta), the company says its team has spent the past year in stealth working with customers and design partners. The release claims Ultra is SOC 2 Type II, ISO 27001, ISO 42001 and AARM Extended conformant, aligns with NIST AI RMF and the Agentic Trust Controls, and has completed a FedRAMP 20x Class B audit with a 3PAO, with FedRAMP certification in progress. Access can be requested at ultra.security. These claims are the company's own and have not been independently verified in the source material.
Go deeper than the headline
Research every firm in this story, track the ones that matter and get the daily briefing — free for 7 days.