Brandkeeper
Nothing off-brand ships.
In the constellation
Brandkeeper highlighted in the live map — hover or tap any node to explore.
The gap
As output scales — and as more of it is drafted by AI — no human can police every artifact for voice and accuracy. Off-brand phrasing and quietly-wrong facts slip through and reach the audience. The tools that exist check tone or check facts; almost none check both together.
The more you publish, the more surface area there is for something off-brand or factually stale to leak out the door.
What Brandkeeper is
Brandkeeper is a gate that makes sure nothing off-brand ships. Every external artifact — a post, an email, a deck, a doc, a support reply — is auto-checked for on-brand voice and factual consistency before it goes out, and gets a clear verdict.
It checks the two things that usually travel separately, together: does this sound like us, and is every claim in it still true? It proposes verdicts rather than silently blocking, and records why on every call — so the gate is auditable and overridable.
What it does
Voice check
Judges every artifact against your brand’s actual voice — does this sound like us, or like someone else?
Fact check with integrity
Verifies that the claims and references in a piece are still true, not just plausible-sounding.
A clear verdict
Renders an on-brand / off-brand call on each artifact before it ships, instead of leaving it to a busy human to eyeball.
Auditable and overridable
Records the reasoning behind every verdict, so calls can be reviewed, trusted, and overridden.
The line between judgment and machinery
AccelMars draws one hard line through every product: what an AI decides, and what runs deterministically. Brandkeeper sits on the boundary — and keeps the two honest.
What the AI judges
- Whether something truly fits the brand’s voice
- How close the tone is to your standard
- Which artifacts need a human’s second look
What stays deterministic
- Verifying every claim against its reference
- Structure and format conformance
- The provenance and trail behind each verdict
Combining a voice judgment with hard reference integrity is the edge: the brand-fit call is the AI’s; the “is this actually true” check is deterministic, not a guess.
Who it’s for
- Marketing and brand teams scaling output without losing the voice
- Orgs publishing a lot of AI-drafted content that has to stay on-brand
- Anyone who needs every external artifact checked before it ships
Planned — designed, not yet built. It builds on the document and integrity engines plus a new brand-voice engine, and shares its voice foundation with Megaphone.