NO SHARED INVENTORY
Models, runtimes, Hermes agents, and tool permissions spread across sites without a common source of truth.
Dafoe gives IT and MSP teams one operator-grade surface to discover, govern, deactivate, fail over, and enforce policy across self-hosted agentic AI runtimes, Hermes agents, models, tools, and workflows.
The waitlist is for teams running local models, internal agents, and tool-calling workflows that need a real command layer before the estate gets wider.
Self-hosted agentic AI gives teams ownership of where intelligence runs and what it can touch. It also creates a live execution surface: agents calling tools, reading local data, chaining workflows, and making decisions inside networks your endpoint, MDM, and observability stack was never built to command.
Models, runtimes, Hermes agents, and tool permissions spread across sites without a common source of truth.
When an agent touches the wrong tool, endpoint, or data path, shutdown is slow, manual, and incomplete.
Runtime and agent failures turn into outages because failover logic is embedded—or missing entirely.
Stop a model, agent, tool, workflow, or entire runtime fleet from one authority point.
$ dafoe disable --scope fleet Move workloads to approved local or remote capacity when health, latency, or policy conditions change.
$ dafoe route --fallback edge-04 Discover and inventory heterogeneous runtimes, agents, models, and tools across offices, data centers, labs, and client sites.
$ dafoe fleet discover Apply desired state and coordinate runtime, model, and tool lifecycle across the fleet from one place.
$ dafoe fleet apply policy.yaml Set boundaries for models, tools, data access, network routes, and execution at runtime.
$ dafoe policy enforce P-104 Dafoe sits above your runtime layer. It does not replace the tools you already operate—it makes them controllable as a system.
Deploy a lightweight Dafoe connector beside each supported runtime or management boundary.
Build a live inventory of models, agents, tools, workflows, endpoints, and dependencies.
Express operating rules as versioned policy: who can run what, where, and under which conditions.
Enforce, reroute, quarantine, or deactivate with a complete record of every action.
Keep agents, inference, and sensitive data inside your environment. Dafoe coordinates the authority layer without forcing a single runtime, model vendor, or deployment pattern.
Dafoe is designed for privileged operations: explicit scope, strong identity, minimal access, and an evidence trail built into every command.
Narrow, explicit permissions at every managed boundary.
Verify the origin and integrity of high-impact control actions.
Maintain evidence for operator actions, policy changes, and runtime events.
Operate across segmented, private, and intermittently connected environments.
Create a shared operating model for Hermes, departmental runtimes, sanctioned models, and internal agent deployments.
Standardize discovery, policy, response, and reporting while preserving tenant and network boundaries.
Revoke tools, quarantine runtimes, disable agents and models, and preserve an auditable chain of action.
Route around unhealthy nodes, coordinate capacity, and understand dependencies before they fail.
Join the Dafoe waitlist for early architecture notes, operator interviews, and access windows for teams actively running self-hosted agentic AI.
Map the runtimes, agents, tools, and shutdown paths you already have.
Preview the command, fallback, policy, and audit surfaces before release.
Get first access when Dafoe opens seats for local AI operators.