Understand how Project KAI is built, operated, and evolved.

The real 8-stage production pipeline -- Status, Source, and Human Control shown for every stage.

The real pipeline flow agents cooperate in, with Human Checkpoint and Planned Connection steps explicitly labeled.

All 13 real agents, grouped into 7 functional layers, each with Capability Status and Autonomous Agent Status.

The real keyword-search tool over 98 real engineering reports -- what it does and doesn't do, stated plainly.

The public system map -- real status cards, the agent network, and quick actions into every real page.

A small, hand-curated, sourced collection of real AI/computing history facts.

A small, curated collection of AI, programming, and robot jokes -- entertainment, not factual intelligence.

Structured AI learning paths -- curriculum architecture only, routing toward the real secondary channel.

Both real YouTube channels, the deterministic content router, and the ecosystem loop connecting them.

Sourced discovery stories and KAI Documentaries -- longer, chapter-based explorations.

Emerging technology signals, every probability a labeled KAI Estimate -- never a guarantee.

Curated, sourced AI and technology news -- not a live feed.

Questions Project KAI is testing, honestly labeled Proposed until actually run.

AI apps and gadgets, honestly labeled by how deeply they were actually evaluated.

An educational, static walkthrough of how a KAI agent is designed to work.

Honest analytics status -- most metrics not connected yet, none fabricated.

Client-side search across the entire Media Network content index.

Bookmarked content -- stored on-device, mirrored to your account when signed in.

A build-time snapshot of real agent, engineering, and production counts -- not a live monitoring dashboard.

Real queue and completion counts for the content pipeline, shown on System Status.

Real, git-sourced engineering history with commit-hash evidence per entry -- not manufactured release notes.

Real internal report categories and engineering metrics -- most remain internal; see what's public below.

How this website's public data is generated, sanitized, and verified from the real KAI OS codebase.

Six real pillars -- Human Review, Engineering Standards, Transparency, Continuous Improvement, Privacy, Future Philosophy.

The real, verified capability-based security model enforced inside KAI OS -- default-deny authorization, kill switches, revocation, rate limiting, audit logging.

The real intent-to-audit pipeline every privileged action flows through, and the honest current status of the Desktop Operator subsystem specifically.

What this static site does and doesn't collect -- true by construction, not just by policy.

Why Project KAI exists and how it relates to KAI OS.

Who's building Project KAI and KAI OS.

Now / Next / Future / Vision -- what's committed and what's direction, kept explicitly separate.

How to reach the project.