Baseline estate posture — routine drift, no active major incident.
6 agents streaming

Autonomy Level Dashboard

Autonomy is a governed property of each use case, not a prompt setting. Elevated-risk actions — refunds, plan changes — trigger a human approval step at the gateway before any downstream dispatch.

The six-level scale

Live population of the estate across levels

L1answer-only

Reads and answers. No side effects of any kind.

1 use cases
L2recommend

Proposes an action for a human to take elsewhere.

1 use cases
L3prepare

Drafts and stages the action; a human commits it.

1 use cases
L4execute-minimal

Executes reversible, single-subject, rate-limited actions.

3 use cases
L5execute-controlled

Executes material actions behind a mandatory approval step.

1 use cases
L6high-autonomy

Executes without per-action approval. Not granted in this estate.

0 use cases

Use cases by autonomy

Each agent's granted level and its risk classification

Use caseAgentLevelRiskApproval gate
Tier-1 service deflection and invoice questionsSalesforce Agentforce — Service CopilotL4 execute-minimalmoderate riskhuman approval
Employee self-service lookups inside TeamsMicrosoft Copilot — Workplace AgentL2 recommendmoderate risknot required
Billing dispute resolution with refund authorityCustom Care OrchestratorL5 execute-controlledhigh riskhuman approval
Autonomous first-line network remediationOpenShift NetOps AgentL4 execute-minimalhigh risknot required
Revenue assurance and reconciliation preparationInternal Finance Analyst AgentL3 preparecritical riskhuman approval
Runbook execution for recurring incidentsInternal Ops Runbook AgentL4 execute-minimalmoderate risknot required
Unreleased prototype under evaluationSandbox Prototype AgentL1 answer-onlylow risknot required

Pending human approvals

Elevated-risk actions held at the gateway

0 in queue

No actions awaiting approval

Trigger an elevated-risk action to see the gate fire.

Call trail

Select any call in the live stream to open its full trail

No call selected

Click a row in the live MCP flow to inspect identity, policy decision, token and result.