Security fixes target the latest v0.1 release and master until a broader support policy is published.

Course modules are untrusted browser code. The runtime bundles them and runs each in a separate iframe with sandbox="allow-scripts", no same-origin privilege, a restrictive CSP, and connect-src 'none'. A minimal validated message protocol carries local events and errors. An iframe failure is isolated from navigation and sibling lessons.

This browser boundary does not make dependency installation or exercise execution safe. Package lifecycle scripts and commands can execute with the learner's local permissions. First-party courses use the committed pnpm lock, an explicit build allowlist, and reviewed dependencies. External compatible courses are unreviewed; inspect their package files, lockfile, modules, and exercise commands before installing or running them. Exercises never run merely because a lesson opens.

The runtime has no accounts, secrets, analytics, backend, or remote learner state. The landing page is static and sets no cookies.

3D model-atlas descriptors are inert data and cannot supply HTML, CSS, shader source, executable code, or runtime URLs. Counts, dimensions, graph references, evidence labels, source references, trace values, and scene budgets are validated before rendering. Renderer resources, observers, listeners, and animation frames are disposed when the explorable is destroyed. WebGL initialization failure or context loss must fall back to the synchronized semantic representation without weakening iframe permissions.

Do not open a public issue for an exploitable vulnerability. Use GitHub's private vulnerability reporting for Doppp/explorables with reproduction, impact, affected commit/version, and any proposed mitigation. If private reporting is unavailable, contact the repository owner through their public GitHub profile and request a private channel.

We will acknowledge a report, reproduce it, coordinate a fix and disclosure, and credit the reporter unless anonymity is requested. Never include real secrets or third-party data in a report.