How much does Zenity cost?

The AWS Marketplace listing for AWS Security Hub Extended is usage-based with no end date: Observability at $130 per resource per month, Runtime Protection at $16 per million tokens per month, and Guardrails for AWS at $2 per million tokens per month, with a 30-day free trial. The main AWS Marketplace listing and direct sales use private offers.

What are the best alternatives to Zenity?

Based on documented capability coverage across the Agentic Index 14 point taxonomy, the closest alternatives to Zenity among security and SOC agents are Arctic Wolf, Horizon3.ai, Okta, Qevlar AI. Each is graded on the same public evidence, and no vendor pays for placement.

Does Zenity support MCP and tool calling?

Full under the Agentic Index verification standard. Zenity publishes public evidence of Model Context Protocol (MCP) support, tool calling, and API or SDK extensibility for connecting agents to external tools. Evidence basis: The Evaluate API lets customers' custom agents call Zenity for inline enforcement decisions on their actions, alongside OpenTelemetry ingestion. It has a public documentation page, Custom Agents: Evaluate API, but the page shows no referenc. Graded from public evidence only against the Agentic Index 14 point capability taxonomy. No vendor pays for placement.

Does Zenity have SOC 2, SSO and role based access control?

Full under the Agentic Index verification standard. Zenity publishes public evidence of enterprise security and identity controls, covering a compliance posture such as SOC 2 or ISO 27001, single sign on (SSO or SAML), and role based access control (RBAC). Evidence basis: Attestations on the trust center cover SOC 2 Type II, ISO 27001 and ISO 27701, with full reports on request. Account setup offers SSO through Azure AD and Okta or generic SAML and OIDC, along with an IP allowlist and encryption keys the cus. Graded from public evidence only against the Agentic Index 14 point capability taxonomy. No vendor pays for placement.

Does Zenity give you tracing and observability into agent runs?

Full under the Agentic Index verification standard. Zenity publishes public evidence of tracing, observability and audit logging into agent runs, so a team can see what an agent did and why. Evidence basis: The customer's agents are analyzed in context: execution paths, tool invocations and memory updates, along with decision flows, RAG queries and file access. Each agent has an activity record, and spans from custom agents come in through Ope. Graded from public evidence only against the Agentic Index 14 point capability taxonomy. No vendor pays for placement.

Back to vendors

Zenity

Q: Who is Zenity for?

enterprise security and platform teams, organizations adopting Copilot, Agentforce, and custom agents, government and public sector agencies, teams governing citizen-developer and shadow AI

Also known as: Zenity Observe, Zenity Govern, Zenity Defend

Visit site

Security / SOC agentindependentVerified 2026-09-28

Entry priceAWS Marketplace (Security Hub Extended): Observability at $130 per resource per month, plus Runtime Protection at $16 and Guardrails for AWS at $2 per million tokens per month.Full pricing detail

Security and governance platform for enterprise AI agents: inventory and execution analysis across SaaS, cloud, endpoint and coding agents, posture checks before go-live, runtime Boundaries that allow, block or shut down agent actions, an Evaluate API and OpenTelemetry for custom agents, and Blue Agent for finding triage with human approval. SOC 2 Type II and ISO 27001.

Zenity is a security and governance platform for the AI agents an enterprise builds and uses. It keeps a live inventory of agents across SaaS platforms, custom and cloud builds, and endpoint and coding agents, and analyzes their execution paths, tool invocations, memory updates, decision flows, RAG queries and file access.

Its capabilities span AI observability, AI security posture management (agent configuration and permissions checked against policy before an agent goes live), exposure management (identity, data and behavior signals correlated into attack paths, with each issue scored 0 to 100), agentic identity and access management, MCP security, and AI detection and response.

Runtime Boundaries evaluate every agent action in real time against the organization's rules and let it through, block it or shut the agent down, in prevention or detection mode, starting from a library of prebuilt rules; a rule written once applies across Copilot, ChatGPT Enterprise, Agentforce, Bedrock, coding agents and homegrown builds. Custom agents connect through OpenTelemetry for visibility and the Evaluate API for inline enforcement.

Blue Agent, the first of Zenity's Guardian Agents, triages findings and suggests a verdict with its reasoning, and nothing is applied without a human's approval. Zenity holds SOC 2 Type II, ISO 27001 and ISO 27701, supports SSO, and is in process for FedRAMP.

Vendor details

Canonical URL

https://zenity.io

Funding status

Independent; founded in 2021 by CEO Ben Kliger and CTO Michael Bargury; a $5M seed (Vertex Ventures, UpWest, November 2021), a $16.5M Series A led by Intel Capital (September 2023), a strategic investment from M12 (July 2024) and a $38M Series B co-led by Third Point Ventures and DTCP (October 2024) (company page).

Company status

independent

Use cases & customers

Primary use cases

AI agent discovery and inventory (shadow AI)AI Security Posture Management and policy governanceintent-aware runtime detection and response for agentsprompt-injection and agent-manipulation defensecitizen-developer and enterprise agent governance

Target customers

enterprise security and platform teamsorganizations adopting Copilot, Agentforce, and custom agentsgovernment and public sector agenciesteams governing citizen-developer and shadow AI

Deployment options

SaaS on AWSfederal deployment within Knox Systems' platform (FedRAMP in process)

Integrations

Covers Microsoft Copilot Studio, Microsoft 365 Copilot, Power Platform, Salesforce Agentforce, AWS Bedrock, Azure AI Foundry, Google Vertex AI, ChatGPT Enterprise, Claude Enterprise and ServiceNow, and since August 2026 GitHub Copilot, OpenAI Codex, Microsoft Foundry and Agent 365, plus any agent emitting OpenTelemetry or gen_ai spans. Custom agents call the Evaluate API for inline enforcement. The docs list a CrowdStrike Falcon SIEM integration, and Zenity is available in AWS Security Hub Extended.

In practice

A security team gets a live inventory of the Copilot Studio, Agentforce and Bedrock agents across the company, with identity, data and behavior signals correlated into attack paths scored by how reachable they are.

A custom agent calls Zenity's Evaluate API before it acts, and Boundaries let the action through, block it or shut the agent down in real time.

Blue Agent triages a new finding in suggest mode and proposes a fix, and nothing is applied until an analyst approves it.

Sources & related URLs

Research sources

https://zenity.io/platform https://finance.yahoo.com/sectors/technology/articles/zenity-named-company-beat-ai-130100277.html https://www.prnewswire.com/news-releases/zenity-expands-ai-agent-security-and-governance-platform-to-include-chatgpt-enterprise-302435218.html https://zenity.io/use-cases/agent-type/home-grown

Agentic Index coverage score

9.0 / 14 capabilities · 64%

Integrations & Tool Calling	Full
Zenity covers Microsoft's Copilot Studio, Microsoft 365 Copilot, Power Platform and Azure AI Foundry, Salesforce Agentforce, AWS Bedrock and Google Vertex AI, ChatGPT Enterprise and Claude Enterprise, and ServiceNow. It also covers GitHub Copilot, OpenAI Codex, Microsoft Foundry and Agent 365. Custom agents report through OpenTelemetry, and Zenity integrates with CrowdStrike Falcon SIEM and ServiceNow. Sourcezenity.io/platformread 2026-09-28
Workflow Orchestration	Partial
Surface, Enforce and Protect run as a fixed product pipeline, and one Guardian Agent is live, Blue Agent, a triage assistant that works in several steps. There is no coordination among several agents and no flow the customer configures. The Clarity and Correlation agents from earlier descriptions no longer appear. Sourcezenity.io/blog/august-2026-product-updatesread 2026-09-28
Knowledge Grounding & RAG	Partial
A live inventory covers agents across SaaS, custom and endpoint deployments, with identity, data and behavior signals correlated into attack paths. There is no retrieval layer that an agent queries, and Blue Agent gathers context for each finding. Sourcezenity.io/platformread 2026-09-28
Human Oversight & Guardrails	Full
Blue Agent runs in suggest mode, so nothing it proposes is applied without a person's approval. On the customer's agents, Boundaries let an action through, block it or shut the agent down. Sourcezenity.io/blog/august-2026-product-updatesread 2026-09-28
Security, Identity & Governance	Full
Attestations on the trust center cover SOC 2 Type II, ISO 27001 and ISO 27701, with full reports on request. Account setup offers SSO through Azure AD and Okta or generic SAML and OIDC, along with an IP allowlist and encryption keys the customer manages. FedRAMP is in process, not authorized. Sourcedocs.zenity.io/ssoread 2026-09-28
Observability & Auditability	Full
The customer's agents are analyzed in context: execution paths, tool invocations and memory updates, along with decision flows, RAG queries and file access. Each agent has an activity record, and spans from custom agents come in through OpenTelemetry, so every tool call the customer's agents make is on record. Sourcezenity.io/platform/ai-observabilityread 2026-09-28
Memory & State Persistence	Partial
Blue Agent's triage guidelines are generated at onboarding and persist for the organization, and the agent suggests revisions when it sees a need. That state persists and the agent maintains it, and Zenity says nothing more about its scope or lifetime. Sourcezenity.io/blog/august-2026-product-updatesread 2026-09-28
Deployment & Data Residency	Partial
Commercial customers get SaaS on AWS. A federal deployment inside Knox Systems' platform is in process for FedRAMP, a second environment not yet authorized. There is no region list or option to host in the customer's own environment. Encryption keys the customer manages protect data but do not set where it lives, and the agent platforms Zenity covers are not where it is hosted. Sourcezenity.io/company-overview/newsroom/company-news/zenity-achieves-fedramp-in-process-status-for-ai-agent-securityread 2026-09-28
Prebuilt Agents, Templates & Packs	Partial
A library of prebuilt Boundaries rules and a snippet library of prebuilt examples ship without a named catalog, and one Guardian Agent, Blue Agent, is live. There is no named set of prebuilt agents. Sourcezenity.io/platform/runtime-boundariesread 2026-09-28
Triggers & Channel Coverage	Full
Boundaries check every agent action in real time as it happens, in prevention or detection mode, and Blue Agent triages findings as they arrive, so events start the work with no one launching it. Sourcezenity.io/platform/runtime-boundariesread 2026-09-28
Model Flexibility & Routing	Not documented
No model provider, choice or routing is named for Blue Agent or the platform's analysis. Zenity covers agents built on many models, which does not give its customers a choice of model for Zenity itself. Sourcezenity.io/platformread 2026-09-28
APIs, SDKs & MCP Extensibility	Full
The Evaluate API lets customers' custom agents call Zenity for inline enforcement decisions on their actions, alongside OpenTelemetry ingestion. It has a public documentation page, Custom Agents: Evaluate API, but the page shows no reference content. Sourcedocs.zenity.io/integration/custom-agents-evaluate-apiread 2026-09-28
Testing, Debugging & Optimization	Partial
Before anything goes live, the customer's agents are checked for configuration and permissions against policy, and attack paths that correlated signals show to be reachable are scored from 0 to 100. That validates configuration. There is no red teaming or scored test run against the customer's agents. Sourcezenity.io/platform/ai-exposure-managementread 2026-09-28
Browser & Computer Use	Not documented
Zenity governs other agents, browser and endpoint agents among them, and does not drive an interface itself or use a browser or computer. Sourcezenity.io/platformread 2026-09-28

The Agentic Index coverage score grades every vendor Full, Partial or Not documented against the same 14 buyer facing capabilities, from public evidence only. Each capability links to how all vendors in the index score on it. How this evidence is graded

Product updates

October 15, 2026·Human approval / guardrailsVerified

Zenity now supports tool call validation in Anthropic's Inference Hooks for Claude Enterprise, so each tool call Claude proposes is held for an allow or deny verdict from Zenity before it runs. Zenity Boundaries can block write, delete and execute operations, calls to MCP servers outside a reviewed list, and actions traced to prompt injection, with one configuration covering claude.ai, Cowork, Claude Code, Claude in Chrome, Claude for Microsoft 365 and Claude Tag. Nothing is installed on endpoints, and customers ask their account team to turn on enforcement.

Bears on: Security / enterprise

View source

August 8, 2026·IntegrationsPartially Verified

Zenity announced an integration with AWS Security Hub to secure enterprise AI agents. The integration extends Zenity's security and governance capabilities directly into the AWS ecosystem.

Bears on: Integrations

View source

August 5, 2026·IntegrationsPartially Verified

Zenity extended its AI agent security and governance platform to support Claude Enterprise. The integration provides visibility, oversight, and risk management for AI agents operating within the Claude ecosystem, including Claude Code and Cowork.

Bears on: Integrations

View source

View all 3 Zenity updates →Tracked since Aug 2026 · Verified from public vendor sources

Pricing

AWS Marketplace (Security Hub Extended): Observability at $130 per resource per month, plus Runtime Protection at $16 and Guardrails for AWS at $2 per million tokens per month.

Per monitored resource per month for observability, plus per million tokens for runtime protection and guardrails (AWS Security Hub Extended); private-offer contracts otherwise.

Trial available

What is public

The AWS Security Hub Extended listing publishes three usage dimensions (Observability $130 per resource per month; Runtime Protection $16 per million tokens per month; Guardrails for AWS $2 per million tokens per month), cancel any time, with a 30-day free trial. The main AWS Marketplace listing is a private-offer placeholder and the vendor site shows no prices.

Variable cost rationale

Charges scale with the number of monitored resources and with token volume through runtime protection and guardrails, with no cap on the usage listing.

Sales call required

Mixed (some tiers require a call)

Free / trial

30-day free trial on the AWS Marketplace listing

Lowest paid plan

Observability, $130 per resource per month (AWS Marketplace, Security Hub Extended).

Key ambiguities

The listing does not define a resource, and whether the Security Hub Extended rates match private-offer pricing for the full platform is not stated.

Agentic Index verified 2026-09-28

Official pricing page Docs

Related vendors

Alternatives to Zenity

The closest documented capability profiles to Zenity among security and SOC agents tracked by Agentic Index, ordered by similarity on the same 14 point evidence the rankings use. No vendor pays for placement.

Arctic Wolf9.5 / 14Adds documented Model Flexibility & Routing
Horizon3.ai8.5 / 14Fuller documented coverage on Deployment & Data Residency
Okta8.5 / 14Fuller documented coverage on Deployment & Data Residency
Qevlar AI8.5 / 14Fuller documented coverage on Deployment & Data Residency
Capsule Security9.0 / 14Fuller documented coverage on Knowledge Grounding & RAG and Testing, Debugging & Optimization
Dropzone AI11.0 / 14Adds documented Model Flexibility & Routing

Similarity is computed from each vendor's Agentic Index coverage score evidence, axis by axis, not from the totals. How this evidence is graded

Compare Zenity with other vendors →Compare all security and SOC agents →See where Zenity ranks →

Data confidence: high

Head to head

Zenity vs DeepKeepZenity vs Noma SecurityZenity vs Pillar SecurityZenity vs StraikerZenity vs HiddenLayer

Zenity

Vendor details

Canonical URL

Category

Funding status

Company status

Use cases & customers

Primary use cases

Target customers

Deployment options

Integrations

In practice

Sources & related URLs

Research sources

Agentic Index coverage score

Product updates

Pricing

Related vendors

Alternatives to Zenity

Head to head

Contact us