Understand what your codebase builds, why it is structured that way, and what a change could affect. One shared Markdown ontology for humans and coding agents—visualized for people, accessible to agents through MCP, and reviewed with Git. Local-first. Open source.
This repository also configures its own agents. See what ontology-atlas tells them →
Agent Claude Code
Coordinator for Atlas product review and the eight design seats. Routes only relevant reviewers, preserves independent evidence, and records the accountable human decision.
Agent Claude Code
Accountable design decider and applier. Reviews real UI evidence, rejects token drift and generic AI styling, prescribes exact changes, edits code, and remeasures the result.
Agent Claude Code
Agent Handoff Designer on the Atlas bench. Keeps real MCP and CLI next actions visible, state-bound, portable, and usable by both people and agents.
Agent Claude Code
Information Visualization Designer on the Atlas bench. Maps every visual mark to a typed ontology fact and measures contrast, graph crossings, density, and colour-independent decoding.
Agent Claude Code
Interaction Designer on the Atlas bench. Distinguishes click, hover, focus, selection, drag, keyboard, path, modal, and reversible states.
Agent Claude Code
Lead Product Designer on the Atlas bench. Converts vague discomfort into one primary moment, one attention winner, and one explicit demotion.
Agent Claude Code
Motion / Action Designer on the Atlas bench. Combines physical feel, interruption continuity, distance-aware timing, frame measurement, and reduced-motion equivalents.
Agent Claude Code
Responsive & Touch Designer on the Atlas bench. Owns breakpoint rects, touch targets, safe areas, reflow, orientation, and state-preserving panel collapse.
Agent Claude Code
Design Systems Engineer on the Atlas bench. Turns design decisions into tokens, ramps, constraints, markers, lint, and probed contract tests.
Agent Claude Code
The shipped product is an installed macOS app. Browser appearance is not desktop proof.
Agent Claude Code
Owner-requested proof auditor for an Atlas product decision. Consumes existing design, responsive, motion, walkthrough, and runtime evidence without repeating those gates.
Agent Claude Code
Evidence reviewer for hard-to-reverse Atlas decisions. Separates observation from inference, names the human failure moment, and prescribes the cheapest proof.
Agent Claude Code
Scope reviewer for hard-to-reverse Atlas commitments. Sets appetite, exposes opportunity cost, and returns the smallest integrated slice with an explicit rollback.
Agent Claude Code
Atlas meaning and sovereignty reviewer. Protects durable evidence-bound meaning, local-first truth, human approval, and next-agent handoff when those boundaries change.
Agent Claude Code
Positioning reviewer for Atlas category, first-contact claims, and one-shot reputation. Tests whether the claim is distinctive, earned, and specific to durable reviewed meaning.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: