Ultra Security
Ultra Security is an early-stage, AI-native security company building an on-host runtime security control plane for the Model Context Protocol (MCP).
Position in SVTR
Snapshot 2026-10-22- Startup SignalsRankedof 2,310new
- Security & ComplianceRankedof 149new
- AgentsRankedof 294new
- AI SecurityRankedof 23
About
Ultra Security is an early-stage, AI-native security company building an on-host runtime security control plane for the Model Context Protocol (MCP). It provides real-time policy enforcement, behavioral analysis and audit logging for every tool call between AI agents and MCP servers, targeting enterprises running multiple agents (Claude, ChatGPT, Cursor, Gemini, etc.) against tool poisoning, credential exfiltration, context injection and unauthorized tool access. It states it is SOC 2 Type II, ISO 27001 and ISO 42001 certified and an early contributor to AARM and the Agentic Trust Control Framework.
Full tier & rationale — an SVTR Intelligence benefit
Capital graph
Funding history
- 2026-10-21Seed$10M
Investors · Ludlow Ventures, Hidden Capital, Entrée Capital, Detroit Ventures, Firedrop, FDVCSource · finance.yahoo.com ↗
Comparable companies· by market
This company
BusinessLatest roundTotal raisedDate- Ultra Security
Ultra Security is an early-stage, AI-native security company building an on-host runtime security control plane for the Model Context Protocol (MCP). It provides real-time policy enforcement, behavioral analysis and audit logging for every tool call between AI agents and MCP servers, targeting enterprises running multiple agents (Claude, ChatGPT, Cursor, Gemini, etc.) against tool poisoning, credential exfiltration, context injection and unauthorized tool access. It states it is SOC 2 Type II, ISO 27001 and ISO 42001 certified and an early contributor to AARM and the Agentic Trust Control Framework.
Seed · $10M · 2026-10
$10M
United States
BusinessLatest roundTotal raisedDate- Asymptote Labs
Runtime telemetry, security detection, policy and identity governance for AI coding agents (Claude Code, Cursor, GitHub Copilot, Codex CLI, etc.); products are the open-source Agent Beacon and the enterprise Asymptote Platform; customers are enterprise security/engineering/platform teams.
Seed · $6M · 2026-10
$6M - Astha.ai
Runtime security scanning, identity and access control, and compliance audit evidence for enterprise AI agent / MCP infrastructure.
— - Certiv
Security control plane for governing and protecting autonomous AI agents inside enterprises
Pre-Seed · $4M · 2026-03
$4M
Europe
BusinessLatest roundTotal raisedDate- Archestra
Data access/modification/sharing control software for securely connecting autonomous AI agents to internal enterprise systems
$14M · 2026-06
$14M - Trent AI
Agentic security solution spanning the agent lifecycle to find and fix vulnerabilities in AI agents and their generated code
Seed · $13M · 2026-04
$13M - NeuralTrust
AI agent security platform for enterprises to discover, monitor and govern autonomous AI agents and block malicious interactions
Seed · $20M · 2026-06
$20M
Other / unspecified
BusinessLatest roundTotal raisedDateConfirmed rounds only. Matched on subsector, sector, stage and funding scale — each verifiable on the company profiles.
Investors that may fit
Accel
21 Security & Compliance deals in the last 24 months · 9 at a nearby stage · 16 in California
Request intro →Y Combinator
21 Security & Compliance deals in the last 24 months · 18 at a nearby stage · 19 in California
Request intro →Andreessen Horowitz
18 Security & Compliance deals in the last 24 months · 9 at a nearby stage · 16 in California
Request intro →Cyberstarts
18 Security & Compliance deals in the last 24 months · 4 at a nearby stage · 14 in California
Request intro →Greylock Partners
14 Security & Compliance deals in the last 24 months · 8 at a nearby stage · 12 in California
Request intro →Ten Eleven Ventures
14 Security & Compliance deals in the last 24 months · 10 at a nearby stage · 7 in California
Request intro →Ranked by each firm’s deals in this sector over the last 24 months, weighted by nearby stage and shared geography. Indicative only — no rating judgment.