Skip to content

Lens Agents#

The governed platform for running AI agents on enterprise systems. Any agent, any model, any environment — your rules.

What is Lens Agents#

Lens Agents gives every AI agent what it needs to get real work done: tools to fetch data, take actions, and manipulate information, and connectivity to the databases, APIs, and systems where business context lives. All governed, auditable, and private.

Lens Agents platform dashboard showing the Audit Trail view — event statistics, activity timeline, and detailed event log for an organization.

See Why Lens Agents for the capability story — what's different, what's shipped, and why it matters for enterprise rollout.

Who is it for#

  • Platform and security teams rolling out AI agents across an organization, who need policy enforcement, audit, and identity for every agent action.
  • Engineers, SREs, and support teams using AI tools day-to-day who want them to reach real enterprise systems without building bespoke integrations.
  • Teams building custom agents (LangChain, CrewAI, Claude Agent SDK, internal frameworks) who need enterprise-grade governance without rebuilding it themselves.
  • Enterprise evaluators assessing AI agent governance as a multi-year procurement decision. Start with Why Lens Agents and the Security section.

  • Getting Started

    How to evaluate Lens Agents and what to expect from a guided engagement.

  • Why Lens Agents

    Consolidated capability story — what's different, what's shipped, why it matters.

  • Core Concepts

    Architecture, request flow, the sandbox, and the protocol agents use — for platform engineers evaluating the approach.

  • Desktop AI Tools

    Claude Desktop, Claude Code, Cursor, Copilot, ChatGPT — governed in minutes, no workflow change.

  • External Agents

    Connect any MCP-compatible agent (LangChain, CrewAI, Claude Agent SDK) with scoped tokens and first-class audit attribution.

  • Managed Agents

    Autonomous agents created on the platform, configured through conversation, operating under your governance.

  • Connections

    Governed access to Kubernetes, AWS, GitHub, and any HTTP system — all audited, credential-isolated, policy-controlled.

  • Administration

    Configure identity, policies, credentials, spending, and audit across your organization — everything platform teams need.

  • Security

    Threat model, sandbox isolation, PII masking, EU AI Act, and the Mirantis compliance framework Lens Agents inherits — the full stack for a CISO security review.

Not sure where to start? Talk to us →

Ready to evaluate?

Every Lens Agents deployment is a guided evaluation today — a hands-on engagement that lands you with working agents, real policies, and your team trained, typically in the first two weeks.

Talk to us to start. See Getting Started for the shape of an engagement.