Records updated
7
Links added
Routes + docs
Secrets exposed
None
Labs content refresh
-> Model report
-> Architecture decision note
-> Embed setup guide
-> Organization training guide
-> RAG architecture note
-> Infrastructure overview
-> Release note
Version
This release note covers the Atina Labs content refresh published on July 14, 2026. It updates the initial starter records with clearer user guidance, sanitized development output, real route references, and more directive technical explanations.Updated Labs Records
- ICL-1 now describes the released capability router, six universal capability types, evaluation output, integration result, and public safety boundary. - Gemini TTS decision is now framed as a neutral architecture decision based on latency and product fit. - Embedding Atina now gives step-by-step setup guidance from Chat, Pricing, and Atina Embed SDK documentation. - Organization AI training now explains where to create an organization, upload knowledge, generate lessons, manage access, learn, and review reports. - Embedded agent RAG architecture now describes the upgraded Layer A/Layer B capability design, RAG sources, live tools, escalation, and analytics path. - Atina AI assistant infrastructure now summarizes the platform capacity across chat, embed, organization training, knowledge, routing, controls, support, and avatar experiences.Product Guidance Added
Users can now follow clickable paths to the right surfaces: Pricing for plans, Chat for the Embed SDK tab, Atina Embed SDK documentation for installation examples, Organization Dashboard for organization management, Organization Knowledge for handbook uploads, Organization Learning for modules, and Organization Reports for progress review.Security And Privacy
The refreshed Labs content intentionally avoids private keys, provider secrets, internal service URLs, raw model artifact paths, exact customer data, and sensitive deployment details. The articles publish architecture contracts, usage steps, and summarized metrics that help users and developers understand the platform without exposing private implementation material.No Breaking Product Change
This migration updates Atina Labs content and adds clickable Markdown link rendering on the Labs article page. It does not change billing behavior, plan enforcement, embed authorization, organization permissions, ICL execution, or SDK runtime behavior.Recommended Reading
Start with Embedding Atina on third-party sites if you are installing the SDK. Start with Organization AI training with company handbooks if you are training a team. Read ICL-1: Intelligent capability routing for embedded Atina agents for the model report.
Related
Continue exploring.
Atina AI assistant infrastructure
A public overview of Atina platform capacity: chat, embed, organization training, knowledge workflows, capability routing, usage controls, analytics, support, and avatar experiences.
Embedded agent RAG architecture
The upgraded Atina Embed architecture: domain context, ICL-1 routing, knowledge retrieval, product feed search, read-only live tools, escalation, response generation, and analytics.
Organization AI training with company handbooks
How organization teams use Atina to turn company handbooks and internal documents into private knowledge bases, training modules, lessons, learner support, access control, and reports.
