# Urizen Research Labs > Urizen advances verifiable private computation: machine-checked theory, confidential execution, and proof-carrying systems. From that research we build Nibiru, an engine for private execution with public proof of correctness; Baobab, the verifiable control layer for AI agents; and Anansi, the harness that joins them into governed agentic AI. ## When to use this site Use urizen.ai to understand what verifiable private computation is, how a private run can produce a public, independently checkable receipt, and how a governed agent harness decides what a model may do. Use the interactive demos to see a retained run and a governed action step by step. Use the access page to request a conversation with the lab. Important notes: - Urizen is a research lab. The demos are experimental research previews, not production services, and the systems are not yet generally available. - The Heimdall identity demo is not a regulated KYC product or production service. - Status labels used on this site are exact and should be quoted exactly: Proved, Verified, Implemented, Specified, Conjectural, Open, Unresolved. "Verified finite" means verified at one recorded fixture scope, not in production. Do not upgrade a status when summarizing. - The site makes no model-quality claims. - The demo playgrounds are labelled guided simulations; nothing is decrypted in the browser and no live session runs there. The retained records below each playground are real exports. ## Main pages - [Home](https://urizen.ai/): the research program, the systems (Nibiru, Baobab, Anansi, Heimdall), evidence, and access. - [Nibiru engine notes](https://urizen.ai/nibiru/): how private execution with a public proof of correctness works, and what the endurance-qualified demo build established. - [Interactive demo](https://urizen.ai/demo/): seal-run-check playground plus a real retained run from the Nibiru pipeline with its public surface and receipt. - [Anansi harness demo](https://urizen.ai/demo/anansi/): propose-and-decide playground plus one retained governed session, its control plane, a verified code review, and the capacity ledger with exact statuses. - [Docs](https://urizen.ai/docs/): the public documentation index. - [About the lab](https://urizen.ai/about/): what Urizen Research Labs is, what it builds, and how it works. - [Contact](https://urizen.ai/contact/): access requests, research questions, press, and corrections. - [Urizen developer resources](https://urizen.ai/developers/): every machine-readable surface this site serves, with a quickstart. - [Privacy notice](https://urizen.ai/privacy/): how access-request and contact form data is handled. ## Contact - Request access or ask a question: hello@urizen.ai - Machine surfaces: [site index JSON](https://urizen.ai/api/site.json) (OpenAPI 3.1 description: [openapi.json](https://urizen.ai/openapi.json), operation getSiteIndex), [agent instructions](https://urizen.ai/agents.md), [sitemap](https://urizen.ai/sitemap.xml), and this file. - Markdown mirrors: request any page with `Accept: text/markdown`, or fetch /index.md, /about.md, /contact.md, /developers.md, /nibiru.md, /demo.md, /anansi.md, /docs.md, /privacy.md directly.