What each is for, where Teleport is strong, and how the two differ, capability by capability.
Teleport · Based on the vendor's public documentation, last reviewed .
Teleport
Teleport is an infrastructure access platform for people and machines. Its MCP access proxies MCP servers, allows or denies each tool by role, records every request, and issues short-lived credentials to users and bots.
Community Edition is free and open source. Enterprise is self-hosted with a paid licence or run by Teleport as Enterprise Cloud. The feature matrix lists MCP access in all three editions.
Vekri
Runtime authorization for AI agents. Vekri Gateway checks every tool call against a verified identity, a task scope and Cedar policy; privileged writes wait for a person's approval of the exact operation; a broker executes with credentials the agent never sees.
In Vekri's cloud, or self-hosted in your Kubernetes cluster with no connection to Vekri.
Side by side
| Capability | What it means | Vekri | Teleport |
|---|---|---|---|
| Decides each tool call inline | Every agent tool call is allowed or denied before it reaches the protected system. | Yes Vekri Gateway decides every MCP tool call sent through it before the broker executes it; a denied call never reaches the system. It is a boundary only when it is the only path to that system. | Documented Teleport proxies MCP traffic and checks each tool call against the user's roles. A denied call is blocked and recorded. [1] [2] [3] |
| Verified workload identity | An agent running as a workload proves who it is with a credential the platform issued, not a shared key. | Yes Kubernetes projected service-account tokens, verified with TokenReview before Vekri issues a short-lived token for the run; every call is then authenticated. SPIFFE X.509-SVIDs from a configured trust bundle (no Workload API yet). | Documented Bots join with platform proof, such as a Kubernetes service account token, and receive short-lived credentials they can use for MCP access. Teleport also issues SPIFFE SVIDs to workloads. [6] [7] [5] |
| Employee agents use company SSO | Agents that employees run sign in as that employee, through the company identity provider. | Yes Vekri is the MCP authorization server for Claude Code and federates to your OpenID Connect provider. SAML sign-in is not available yet. | Documented People sign in with tsh, through single sign-on, and their MCP clients use that session. Enterprise editions support OIDC, SAML and GitHub; Community Edition supports GitHub. [8] [9] |
| Task-scoped permissions from the server | What an agent may do is set per task by the server, not by the agent or its prompt. | Yes A run scope for workloads, or a protected access assignment for employees. The effective permission is always an intersection, never a widening. | Partly Just-in-time Access Requests grant extra roles for a limited time, typically for a whole MCP server. A scope issued by the server for each agent task is not described. [10] [9] |
| Policy as code, deny by default | Decisions come from versioned policy, and anything not allowed is refused. | Yes Cedar over typed operations. Unknown operations and evaluation errors are denied, and a revision can be simulated before it is published. | Documented Roles list allowed and denied MCP tools by name, glob or regular expression. No tool is allowed unless listed, deny takes precedence, and rules on tool arguments are not described. [2] |
| Approval of the exact operation | A person approves one specific operation, and the approval cannot be reused for another. | Yes Privileged writes wait for another person to approve that operation, bound to its digest, short-lived and single use, with separation of duties. | Partly Reviewers approve Access Requests, which grant access to an MCP server for a period, often hours. Approval of a single tool call is not described. [10] |
| Agent never holds backend credentials | Secrets for the protected system stay out of the agent's reach. | Yes A broker executes with narrowly scoped Kubernetes or GitHub App credentials. Incoming tokens are never forwarded. | Documented Clients hold short-lived Teleport credentials, and Teleport sends the MCP server its own signed JWT identifying the user and roles. [4] [7] |
| Audit evidence of decision and outcome | Each call leaves a durable record of what was decided and what happened. | Yes Intent is recorded durably before execution and the outcome after; an uncertain outcome is never retried blindly. | Documented Each MCP request, including denied ones, is recorded as an audit event carrying the JSON-RPC message, with the tool name and arguments. [3] [9] |
| AI agents on workstations | Finds AI coding agents and other AI applications on employee devices, and watches or controls what they do there. | Partly Vekri Endpoint discovers coding agents, desktop and IDE AI apps, local models and their MCP servers on macOS and Linux, and its local MCP gateway decides each MCP tool call sent through it (Claude Code verified connecting; tool calls verified with an MCP client library). On macOS, a network gateway mediates MCP requests to approved remote servers, and agent control stops AI agents the organization has not allowed. Hooks observe built-in tools without blocking them; Windows is not supported. | Not in public docs Agent visibility and discovery is listed as in progress in Teleport's Agentic Identity Framework. Discovering coding agents on workstations is not described as available. [11] |
| Prompt and output inspection | Detects prompt injection or sensitive data in what models read and write. | Partly Tool results on Vekri Gateway and the local MCP gateway are inspected for prompt injection before the agent reads them; sensitive-data categories are detected on the device, off by default. Browsers and file uploads are not inspected. | Not in public docs Inspecting prompts or tool output is not described in the MCP access documentation. [1] |
| Systems covered | Which systems, platforms or agents the product works with. | Kubernetes and GitHub, with GitLab, PostgreSQL, MySQL, MongoDB, Slack, Jira, Linear, Asana, ClickUp, monday.com, HubSpot, Pipedrive, Zendesk, Freshdesk and custom OpenAPI connectors in preview. | Servers, Kubernetes clusters, databases, applications and MCP servers over stdio, SSE or streamable HTTP, used from clients that tsh configures, such as Claude Desktop and Cursor. |
| Deployment | How the product is delivered. | In Vekri's cloud, or self-hosted in your Kubernetes cluster with no connection to Vekri. | Community Edition is free and open source. Enterprise is self-hosted with a paid licence or run by Teleport as Enterprise Cloud. The feature matrix lists MCP access in all three editions. |
Every agent tool call is allowed or denied before it reaches the protected system.
An agent running as a workload proves who it is with a credential the platform issued, not a shared key.
Agents that employees run sign in as that employee, through the company identity provider.
What an agent may do is set per task by the server, not by the agent or its prompt.
Decisions come from versioned policy, and anything not allowed is refused.
A person approves one specific operation, and the approval cannot be reused for another.
Secrets for the protected system stay out of the agent's reach.
Each call leaves a durable record of what was decided and what happened.
Finds AI coding agents and other AI applications on employee devices, and watches or controls what they do there.
Detects prompt injection or sensitive data in what models read and write.
Which systems, platforms or agents the product works with.
How the product is delivered.
More comparisons
Aembit
Aembit is an identity and access platform for workloads and AI agents. Edge components inject short-lived credentials into workload requests, and its MCP Identity Gateway applies access policy and tool rules to every MCP request.
Microsoft
Microsoft Entra Agent ID gives AI agents their own identities in Entra, created from blueprints, and extends Conditional Access, ID Protection and governance to them. Access is decided when Entra issues a token.
Okta, Inc.
Okta for AI Agents registers agents as identities in Okta's directory and controls which resources they may reach. Agent Gateway, generally available since October 2026, proxies remote MCP servers and injects credentials into each tool call.
Delinea
StrongDM, now part of Delinea, proxies access to databases, servers, Kubernetes, clouds and MCP servers through nodes in your network. Cedar policy can allow or forbid individual MCP tools, and access can require approval.
Contact Sales
Talk to us about the AI your organization uses, the systems agents reach and how you deploy: in Vekri's cloud, or inside your own Kubernetes cluster.