This page is not the Loom landing page and not the whole product thesis. It shows how Loom + Kernel look on the live host, then shows the first-party Meridian workflows that currently prove the runtime under real pressure.
These charts read directly from the live proof routes on this host. They stay here so the runtime picture remains visible while you scroll through the first-party workflow examples.
Loading runtime footprint…
Loading queue state…
Loading proof status…
The workflow example here is intentionally commercial because it pressures the runtime in a useful way. But treat it as a first-party app on Loom, not as Loom's identity.
OpenAI, Anthropic, Google.
Models, pricing, policy, deprecations, and runtime risk.
The workflow stays inside Loom + Kernel instead of loose docs and chat.
Pricing shifts, new models, forced migrations, and API deprecations become inputs to a governed signal stream instead of random bookmarks.
Context changes, Claude launches, and safety posture changes become cited inputs instead of rumor-driven updates.
Gemini pricing, preview shutdowns, and Vertex changes land as governed runtime artifacts rather than scattered notes.
This is a shortened reference slice. The point is to show how a first-party Loom workload turns signal into operator-ready shape without turning this page into a wall of prose.
One governed stream becomes two different first-party outputs: a calmer leadership brief and a sharper GTM asset. That reuse matters because it proves Loom can back real operator workflows instead of isolated toy prompts.
If you want the runtime itself, go to /loom. If you want the truth line for all claims, read the boundary note.
Want the runtime instead of the example?
Install Loom Compare Loom vs the field See the current first-party pilot path
Meridian portal for Loom, Kernel, and first-party apps © 2026