Skip to content

Surfaces

One declaration in the repository becomes the command line, the HTTP API, MCP, the UI, the Cockpit, the documentation and this site, and a projection that disagrees with it fails the build.

The failure it answers

Interfaces disagree. The command line, the API, the docs and the website each keep a copy, and the copies drift.

How the repository's declarations reach the people and tools that read them: a capability declared once and projected to every interface, majordomus ui that shows a person how the delivery stands, the Cockpit beneath it that renders the registry, the installer, and the catalogue of failure modes this product answers. None of the interfaces is maintained by hand, and a projection that has drifted from its declaration is a build failure rather than a support ticket.

Not in this domain: the rules those interfaces carry, which is governance; and what is true about them, which is evidence.

The features filed under it

Each feature names this domain in its own file; the domain lists none of them. A feature added with this domain is on this page at the next build, and one removed is gone.

  • Declare once

    A capability is written in one place, and the command line, the HTTP API, the OpenAPI document, MCP, the Cockpit, the reference and this website are projections of it.

    9 of 9 claims guaranteed with a named test · 1 use case

  • Interfaces

    Every agent, script and person reads the same repository through the interface it already speaks, and none of the interfaces is maintained by hand.

    11 of 11 claims guaranteed with a named test · 3 use cases

  • Observatory

    One screen of the UI rotates eight scenes on a display nobody is sitting at, and every mark on it is a value a capability answered; a scene with nothing true to draw says why and is never drawn empty.

    1 of 1 claims guaranteed with a named test

  • UI

    The shared server's primary surface answers how the delivery stands — health, releases, the roadmap, the work and its pull requests, the mesh, the evidence and the rules — and every value on it names the capability it was read from and every other surface that serves it.

    1 of 1 claims guaranteed with a named test

  • Cockpit

    The same server that answers agents renders every capability, object, graph, worktree and health check as pages, with a runner generated from each capability's own schema; nothing on it is a second list.

    1 of 1 claims guaranteed with a named test

  • Install

    The installer downloads one archive for your machine, verifies its digest against the release metadata before unpacking anything, and installs atomically; no git, no Rust, no Node, no root.

    9 of 9 claims guaranteed with a named test · 5 use cases

  • Why

    Every operational moment the tool exists for is one file with the signals that identify it and the mechanisms that answer it, and a reader's own symptoms resolve to a diagnosis by counting, not by inference.

    3 of 3 claims guaranteed with a named test · 1 use case

The failure modes its features answer

Operational moments from the Why catalogue that name a rule, a command, a capability or a claim of a feature in this domain. Nothing here was assigned.

Every failure mode it answers (38)

What is claimed

Every claim a feature of this domain names, once, with the status it declares — guaranteed, advisory, planned or rejected — and the test that settles it. Whether a recorded run supports a claim is its own page's verdict.

35 claims, 35 guaranteed with a named test

Executable use cases

Scenarios the tool runs against itself every time this site is built.

Searches the titles and descriptions of every page this site publishes.