Noma Security

What it does

Noma Security is an enterprise platform for AI agent security and governance. It takes in agent activity across endpoint, SaaS, and cloud environments, including agents, MCP servers, tools, skills, models, prompts, tool calls, and responses. It produces inventories of AI assets, risk findings and posture assessments, runtime allow or block decisions, masked data, alerts, human-approval routing, and audit trails with framework mappings for review.

Security teams use the platform in three main ways: discovering agents and AI components their organization has not approved, testing AI applications and agents with automated multi-turn adversarial campaigns, and enforcing per-agent and per-tool policies at runtime. Its distinguishing approach is to judge each action against session context, the identity the agent acts for, reachable data, and baseline behavior, rather than only blocking access.

The platform suits enterprise security and governance teams that need agent visibility and control. Access starts with a demo request on the vendor's site, and buyers should confirm pricing with the vendor. Deployment relies on existing infrastructure such as agent hooks, gateways, EDR/MDM, and CI/CD pipelines, so coverage depends on those integration points.

What it can do

• Discover AI agents, MCP servers, models, and skills across endpoint, SaaS, and cloud environments
• Assess posture and identify toxic combinations of agent capabilities using out-of-the-box policies
• Enforce per-agent and per-tool access policies for MCP connections, tools, and actions
• Run automated multi-turn adversarial red-team campaigns against AI applications and agents
• Detect prompt injection, data leakage, tool poisoning, and rogue behavior, with actions such as block, mask, steer, human approval, or killswitch
• Map findings to NIST AI RMF, EU AI Act, ISO 42001, OWASP LLM Top 10, and MITRE ATLAS for audit evidence

Works with

• Cloud platforms
• SaaS platforms
• Code repositories
• EDR and MDM
• CI/CD pipelines
• AI and MCP gateways
• Agent hooks
• Direct APIs

Before you choose

• Pricing, trial availability, and self-serve options are not published; access starts with a demo request, so buyers should confirm cost with the vendor.
• Runtime enforcement depends on deploying through existing agent hooks, gateways, or EDR/MDM, so coverage depends on those integrations.
• Detection accuracy and false-positive reduction are vendor statements and should be validated in a proof of concept.

Listed under

• Security
• AI and agent security
• AI agents
• AI security
• Compliance
• Enterprise security
• Governance
• Red teaming

Get started

The primary call to action on the site is a demo request.

1. Connect cloud platforms, SaaS platforms, code repositories, and endpoints through existing EDR or MDM so AI assets can be discovered.
2. Route agent traffic through supported hooks, AI or MCP gateways, or direct APIs so runtime policies can be applied.
3. Apply out-of-the-box policies and protection profiles, then tune individual detectors and add per-agent or per-tool rules.

Pricing and plans are arranged with the vendor; request a demo to confirm cost and billing model.

We couldn’t find published prices for Noma Security.

Sources for this overview

Product pages, documentation, and pricing checked 29 Oct 2026. Each source supports part of the overview; availability and prices can change.

• Agent Security and Governance | Noma ↗noma.security
• AI Red Teaming: Automated Adversarial Testing | Noma ↗noma.security
• AI-SPM: AI Security Posture Management | Noma ↗noma.security
• AI-DR: AI Detection and Response | Noma ↗noma.security

Company website

noma.security: product, pricing and policy pages

YC directory mirror

Batch and company name

Web search

Confirms the official site

Social accounts

Accounts the company links, and its posts on X

Human review

Checked by a person

knowledge_base

manual_review

Last checked 29 Oct 2026. Every field keeps its source and date; unknown means we couldn’t confirm it.