Agent workflows become useful in operations only when they are governed. Teams need boundaries for what agents can use, which tools can execute, and how runtime behavior is reviewed.
Dekaplane positions agents inside the control plane instead of leaving them as disconnected scripts or unmanaged bots.
The operational agent problem
Uncontrolled agents can create risk: unclear tool access, missing traces, inconsistent model routes, and no shared administrative view. A self-hosted agent platform gives operators a stronger foundation.
Bound tool execution
Visible workflow runs
Admin-selected model routes
Clear separation between portal access and local runtime access
Dekaplane's agent model
Dekaplane connects agent workflows to the same control-plane configuration used for ChatOps, tools, MCP integrations, and gateway routing. The result is a platform surface operators can reason about.
Configure tools before use
Coordinate ChatOps and agent workflows
Track execution state and trace events
Use the AI Gateway for model access
Designed for self-hosted evaluation
The current package is intentionally focused: one self-hosted tenant, local login, release download, license file, and documentation aligned to the available release.
Dekaplane is currently available as a Free single-tenant self-hosted release with portal-based downloads,
license materials, and installation guidance.