Skip to main content
The Agents Hub is one place in the Endor Labs user interface to discover the security agents Endor Labs publishes, understand what each one does, and see how your team is using them. Each agent packages a specific Endor Labs workflow, such as triaging findings, remediating vulnerable dependencies, or diagnosing scan failures. You run the agents inside your own AI coding assistant, and the Agents Hub gives you the catalog and the activity view in the product.

​
How it relates to the Agent Kit

The Agents Hub and the Endor Labs Agent Kit describe the same agents from two directions. The agents are read-only against Endor Labs data. When an agent proposes a change, that change lands as a pull request or merge request in your source control provider, never as a direct write to Endor Labs. See Trust and attribution for the full model.

​
Open the Agents Hub

Select Agents Hub from the left sidebar. The catalog lists every published agent, with an Active agents section for the agents your team has used recently. You can filter the catalog by category or search it by name. Select any agent to see what it does and which AI coding assistants it supports. See Browse agents for a full tour of the catalog, the agent detail view, and the catalog states.

​
Run an agent

The agents run inside your AI coding assistant, not in the browser. To start using an agent, install the Agent Kit in your host, then invoke the agent through that host. When an agent reaches Endor Labs data during a run, it authenticates with a short-lived, read-only credential that Endor Labs attributes to you. You don’t hand the agent your API key. See Delegated agent credentials for how this works and what to set up.

​
Monitor usage

The Agents Hub shows you how your team uses each agent, including its Tasks count and Last run time. This view is usage metering, not a compliance audit log. See Agent activity for what the numbers mean and how to read them.

​
Agents in the catalog

The Agents Hub publishes these agents. Each one shows its category in the catalog, and you can filter on that category to narrow the list. For how each agent installs into a host, see the Agent Kit.

​
Availability

Requires endorctl v1.7.1088 or later. Endor Labs authors and signs the catalog. The agents in the Agents Hub are designed to be implemented in your AI SDLC and workflows. You can view the agents available for implementation in the Agents Hub, but you can’t yet add a custom agent inside Endor Labs. To see the available agents in more detail, and to learn how to create your own custom agent, see the open source Endor Labs Agent Kit.

​
Explore the Agents Hub

Browse agents

Find an agent in the catalog, read what it does, and install it in your AI coding assistant.

Agent activity

Read the usage metrics the Agents Hub reports for each agent in your tenant.

Delegated credentials

See how an agent reaches Endor Labs data under a short-lived, read-only token attributed to you.

Trust and attribution

Understand how Endor Labs signs the catalog and attributes every agent action to a person.