Chat, Skills, and Riverbed IQ Ops Assist
Riverbed IQ Ops offers more than one AI
Artificial Intelligence. Technology that applies intelligence and automation to observe environments, detect anomalous behavior, and execute automated investigations.-powered path. This topic explains how Chat with Riverbed Q (Q), the Skill Catalog, Agents, and the Observation Feed relate to Riverbed IQ Ops Assist and other chat experiences in the console.
Chat, skills, agents, and observation feed
These are the AI-assisted operations capabilities introduced in AI-Assisted Operations in Riverbed IQ Ops:
-
Chat with Riverbed Q (Q) or a published catalog agent on the Chat page and embedded assistant surfaces.
-
The Skill Catalog for skills and commands.
-
The Agent Catalog for catalog agents (open from Automation > AI Agents). Agents run in the Agentic AI runbook node. To run an agent on a timetable, schedule an on-demand runbook that includes that node. See Agents.
-
The Observation Feed widget for actionable observations.
They share the Skill Catalog, Agent Catalog, and chat gateway services aimed at conversational design, governed reusable content, and runbook execution.
Riverbed IQ Ops Assist and the GenAI node
Riverbed IQ Ops Assist is the established optional feature you enable from Riverbed IQ Ops Assist Configuration. When Assist is enabled, the GenAI runbook node calls the Generative AI language model with instructions and queries you configure in the node editor, including inference settings.
Use Assist and GenAI when you need direct LLM inference in a runbook with the Assist enablement and data-region policies described on the configuration page.
When to use which
|
Goal |
Suggested starting point |
|---|---|
|
Explore data, draft runbooks, or get interactive help |
|
|
Publish reusable instructions for the assistant and automations |
|
|
Author catalog agents for runbook automation and scheduled on-demand runbooks |
|
|
Run a catalog agent inside a runbook branch |
Agentic AI node |
|
Call a configured LLM with explicit prompts in a runbook |
GenAI node with Riverbed IQ Ops Assist enabled |
|
Triage observations that need operator action |
Subprocessor inference locations
When Chat, agents, and other AI-assisted operations features are enabled, Riverbed leverages large language model (LLM) inference services provided by its subprocessors.
AWS (Amazon Bedrock): Customer Data remains stored exclusively in AWS data centers in the customer-selected data center region. During inference, requests may be routed to other AWS regions within the same geography (e.g., NA, EU, or APAC) to optimize performance and availability. All data transmitted between regions remains on AWS's private network backbone and is encrypted in transit. It does not traverse the public internet. No Customer Data is stored outside the customer's selected region.
The table below maps the customer-selected AWS data center region to the corresponding inference processing locations for AWS. This table is a starting point and may be updated as additional regions and models are finalized.
|
Customer-Selected AWS Data Center Region |
Model Processing Locations |
|---|---|
|
Australia |
Australia India Japan Singapore South Korea |
|
Canada |
Canada |
|
Germany |
France Germany Ireland Sweden |
|
Singapore |
Australia India Japan Singapore South Korea |
|
Sweden |
France Germany Ireland Sweden |
|
United Kingdom |
France Germany Ireland Sweden United Kingdom |
|
United States |
United States |
Other chat in the console
Data Forensics may expose a legacy AI chat panel when that feature is enabled for Navigator. That panel is not Chat with Riverbed Q (Q): it uses different enablement, does not use the Skill Catalog slash palette, and does not share the same Orion chat terms acceptance flow. Click Q in the left menu for the assistant experience described in Riverbed IQ Ops Chat.
For Trust Center and AI-in-products policies that apply to Assist, see the link on Riverbed IQ Ops Assist Configuration.
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".