Creating and Editing Agents

Use the agent editor to define catalog agents that Riverbed IQ Ops can run inside runbooksClosed An automated workflow that executes a series of steps or tasks in response to a triggered event, such as the detection of anomalous behavior generating an incident, a lifecycle event, or a manually executed runbook.. The editor includes metadata fields, a Markdown instructions area, linked skills, and an embedded assistant so Riverbed Q (Q) can help while you write.

To browse the catalog, see Agents. For runbook execution, see Agentic AI. For linked skills, see Skills and Commands.

Open the editor

Create a new agent

To create a new agent:

  1. To open the Agent Catalog:

    1. Click Automation in the left menu.

    2. Click AI Agents.

  2. Click + New Agent.

The page title is Create New Agent.

Edit an existing agent

To edit an existing agent, open its card from the Agent Catalog. The page title is Edit Agent.

Fields

  • NAME and DESCRIPTION: Display name and short summary of what the agent does.

  • VERSION: Version identifier for the agent. This field is read-only while you edit an existing agent.

  • SKILLS: Skills from the Skill Catalog that the agent can use. Select one or more published skills, add a custom label in the picker, or click Add skill to open the skill editor and create a new entry.

  • AGENT INSTRUCTIONS (MARKDOWN): The agent behavior, goals, and constraints. Use the Preview and Edit tabs to review rendering. New agents open with starter template content you can replace.

Status, save, and publish

Use the status control in the page header to set lifecycle state:

  • Draft: Work in progress that is not yet available for runbook use.

  • Published: Ready for use in automation.

  • Disabled: Temporarily unavailable without removing the agent from the catalog.

  • Archived: Retired from active use.

Click Save Agent to persist your work. You must select at least one skill before you can save. After a successful save, the page returns to the Agent Catalog. Click Discard to leave without saving. Click Back to Catalog to return to the catalog list. Use Test Agent when you want to try the agent from the editor before you rely on it in a runbook.

Note: If save fails, the UI shows an error message. Correct validation issues (for example, a missing name) and try again.

Embedded assistant

When enabled in your environment, an assistant panel appears beside the editor so you can ask Q for help while you draft agent instructions. The panel uses the same chat terms and sessions as standalone Chat. See Embedded AI Assistant.

Create a runbook from an agent

Click Create Agent Runbook to open on-demand runbook creation with an Agentic AI node already set to the current agent. Use this shortcut when you want a runbook that calls the agent as part of the flow, then save and publish the runbook like any other on-demand runbook.

View and manage schedules

Note: Schedules apply to on-demand runbooks, not to the agent catalog entry itself. To run an agent on a timetable, schedule the on-demand runbook that contains the Agentic AI node for that agent.

Click View Schedules in the agent editor to open the Runbook Scheduling page. From there, create or edit schedules for the on-demand runbooks that call your agents. You can also open Scheduling from the left menu:

To open the Runbook Scheduling page:

  1. Click Automation in the left menu.

  2. Click Scheduling.

AI-Generation disclaimer

When using Generative AI technologies to create and generate content, the target audience must be aware that the content is AI generated. We recommend adding a disclaimer to the output of the content generated using the GenAI node. For example, appending "AI-Generated content".