Multi-Agent Workflows

Overview

When you ask a question in AI Chat, OpsWorker can invoke multiple AI agents to gather information from different sources. Agents are selected automatically based on your question and the integrations configured for the cluster.

How It Works

flowchart TD
    Q[Your question] --> Router[Agent Router]
    Router --> K8s[Kubernetes Agent]
    Router --> Git[Git Agent]
    Router --> DD[Datadog Agent]
    Router --> Obs[Observability Agent]
    K8s --> Combine[Combine results]
    Git --> Combine
    DD --> Combine
    Obs --> Combine
    Combine --> Answer[Answer with full context]

Automatic Selection

OpsWorker determines which agents to invoke based on:

• Question content — "Show me logs" → Kubernetes Agent; "What PRs were merged?" → Git Agent
• Available integrations — Only agents with configured integrations are considered
• Investigation context — If chatting about a specific investigation, agents focus on relevant resources

Example: Multi-Source Investigation

Question: "Why is the api-gateway service slow?"

OpsWorker may invoke:

1. Kubernetes Agent — Checks pod status, resource utilization, events, logs for error patterns
2. Observability Agent — Runs PromQL via Grafana MCP for latency percentiles and request rates, searches Loki logs for error patterns, checks relevant dashboards
3. Git Agent — Checks recent deployments and code changes that may have introduced a regression

The response synthesizes data from all sources into a coherent answer.

Agent Transparency

The portal shows which agents were invoked during each response:

• Execution blocks — Each agent's activity is displayed, showing what tools were used and what data was retrieved
• Tool calls — See the specific queries each agent ran (e.g., "get pods in namespace production")

This transparency helps you understand how the AI arrived at its answer and verify the data sources.

Agent Availability

Agent	Required Integration	Enabled By	Capabilities
Kubernetes Agent	Kubernetes Agent installed	Always (core requirement)	Pods, services, logs, events, configs
Git Agent	GitHub or GitLab	GitHub or GitLab integration	Commits, PRs, blame, file contents
Datadog Agent	Datadog	Datadog integration	Metrics, traces, monitors
Observability Agent	Grafana MCP	Grafana MCP integration	PromQL metrics, Loki log search, dashboards, alert rules, incidents, on-call schedules

note

The Observability Agent requires the Grafana MCP integration (OpsWorker → Grafana query access). This is separate from Grafana Alerting (Grafana → OpsWorker alert webhook). See the Grafana Integration guide for details on both.

Context Sharing

Within a conversation, agents share context:

• Data gathered by one agent is available to others in the same session
• Follow-up questions build on previous answers
• The AI maintains awareness of which resources and namespaces have been discussed

Next Steps

• AI Agents Overview — Learn about each agent
• Example Prompts — Try multi-agent prompts