Dekaplane centralizes model access, AI Gateway configuration, governed ChatOps, agent workflows, tools, and observability in one customer-operated platform.
An AI control plane should do more than list models. Operators need a place to govern access, connect tools, understand workflow behavior, and move from configuration to safe action.
Dekaplane is built around that operating surface: model and provider onboarding, AI Gateway management, ChatOps policy, agent workflow coordination, and traceable runtime behavior.
What problem this solves
Teams adopting LLM-backed operations often end up with separate routing config, ad hoc Slack bots, unmanaged tools, and limited visibility into what happened. Dekaplane gives those concerns one administrative home.
Model and provider access are configured centrally
Workflow behavior is governed before operators use it
Tool and MCP access can be reviewed as part of the platform
Runtime events remain visible for debugging and audit
How Dekaplane approaches it
Dekaplane treats the control plane as the source of operational intent. Admins configure providers, aliases, tools, ChatOps settings, and workflow policy; the runtime follows that configuration instead of relying on scattered manual setup.
Unified model/provider onboarding
Managed gateway aliases and verification
Agent and ChatOps policy surfaces
Observability across requests, workflows, and tools
Why self-hosted matters
For infrastructure and operations teams, sensitive integrations, credentials, workflow state, and debugging records are often better kept inside the customer environment. Dekaplane is packaged for that self-hosted operating model.
Customer-operated runtime
Local tenant-admin inside the installed platform
Portal-based release and license delivery
Current public offer: Free single-tenant self-hosted
Dekaplane is currently available as a Free single-tenant self-hosted release with portal-based downloads,
license materials, and installation guidance.