Noma Agent Access Control
Builds a live registry of allowed agents, MCP servers and skills, scopes which actions each agent may take by its owner's identity, and blocks actions that drift from the agent's intended task.
Find alternatives to Noma Agent Access Control
What it does
- Guardrails and safety
Sources
- Pricing
-
Product pages offer "Get a Demo" only; no rate is published on the product page or the request-demo page, so models are left empty.
- Description
-
Official documentation · 16 Nov 2026
Page: "Define what actions each agent can take, scope permissions by identity, and enforce policies at the moment a connection or action is attempted."
- What it does
-
Official documentation · 16 Nov 2026
Page: "Noma compares attempted actions with the agent’s intent to detect drift and prevent excessive actions." and "scope permissions by identity, and enforce policies at the moment a connection or action is attempted." The page never mentions credentials, secrets or tokens, so agent-credential-management (holding secrets so the agent never sees them) is not claimed.
- Sold within
-
Product page presents it inside "The Noma Agent Security & Governance Platform" with no packaging statement for this module, so sold_within is recorded as not stated.
- Name
-
Official documentation · 16 Nov 2026
Page title "AI Agent Access Control | Noma"; Products menu "Agent Access Control".
- URL
-
Official documentation · 16 Nov 2026
Product page read 2026-11-16 (server-rendered); listed under Products as "Agent Access Control".
- Deployment
-
Official documentation · 16 Nov 2026
Page: "Noma applies your policies through supported gateways, hooks, and APIs"; no self-hosted option is stated.
Also from Noma
-
Noma AI Red Teaming Platform
Noma AI Red Teaming runs multi-turn attack campaigns against AI applications and agents and maps the findings to OWASP LLM Top 10, MITRE ATLAS, NIST AI RMF, the EU AI Act and ISO 42001.
-
Noma AI-SPM Platform
Finds every AI agent, MCP server, tool, skill and model across endpoint, SaaS and cloud, scores the risk of each agent's combined tools and data access, and maps findings to AI security frameworks.
-
Noma AI-DR Platform
Detects and stops dangerous or drifting AI agent behaviour at runtime using session context, identity and intent, with block, mask, steer, human-in-the-loop or kill-switch responses.
Something wrong here? Send a correction — quote this product id: noma-security-noma-agent-access-control.