Skip to content
Planned Knowledge engine

Prism

One structured source of truth, rendered into slides, sites, reports, and docs — presentation as a deterministic transform.

Replaces rebuilding the same content by hand in five formats

In the constellation

Prism highlighted in the live map — hover or tap any node to explore.

01Why

Why it exists

The same content gets rebuilt by hand in five formats — a deck, a microsite, a PDF report, a doc — and the moment the source changes, every copy drifts out of sync. Presentation is endless manual rework on top of facts you already have.

No engine owns the step from structured truth to a polished, publishable artifact, so each app reinvents rendering and each format is maintained on its own.

02What it is

What Prism is

Prism takes one structured source of truth and renders it into slides, sites, reports, and documents — many output formats refracted from a single input, reproducibly.

It renders; it doesn’t author. Design taste lives in templates and in AI-assisted layout proposals you approve, while what to say and what to emphasize stays with the app. Change the source once and every format re-renders true to it.

03Capabilities

What it does

One source, many formats

A single structured model rendered into slides, sites, reports, and documents.

Template binding

Named templates carry layout and structure; the same model and template always produce the same artifact.

AI-assisted layout

Layout and visual hierarchy proposed with AI help, captured in templates a human approves — not improvised per render.

On-brand styling

Applies brand styling supplied by the voice engine; embeds assets with provenance intact.

04The cleavage

The line between judgment and machinery

AccelMars draws one hard line through every product: what an AI decides, and what runs deterministically. Prism sits on the boundary — and keeps the two honest.

AI judgment

What the AI does

  • Propose layout and visual hierarchy
  • Help shape templates from a model
  • Suggest format-appropriate presentation
Deterministic

What stays deterministic

  • Render of a model + template to a format
  • Reproducible output across targets
  • Asset embedding with provenance

Prism renders, it doesn’t author. AI assists with layout and gets approved into templates; the render itself is deterministic — the same model, template, and style produce the same artifact every time.

05Connections

How it connects

Powered by

Engines Prism builds on.

Used by

Apps composing Prism.

Planned

Planned — designed, not yet built. It will power rendered reports in Keystone, multi-format publishing in Megaphone, and knowledge artifacts in Ops and Atlas.

Open-core, like Anchor and Booster — the mechanical engine in the open.