> ## Documentation Index
> Fetch the complete documentation index at: https://docs.omni.co/llms.txt
> Use this file to discover all available pages before exploring further.

# November 13, 2026

> Apps Performance Inspector, MCP Links Back to Omni, Modeled Window Functions, Per User LLM Restrictions, Automated Learning (Including End User Suggestions), and the Launch of Python

## Automatic Dashboard Formatting

*`Richard Czechowski · Dashboards UX`*

This demo shows how a single tidy-up action can apply formatting across an entire dashboard. The system uses field names and chart metadata to interpret whether changes are positive or negative, adjust KPI colors, and apply conditional formatting and alignment to tables.

[https://www.youtube.com/embed/4kEn30lZIho](https://www.youtube.com/embed/4kEn30lZIho)

## Filter-Aware Blobby on Dashboards

*`Ben Horne · Dashboards Workbook`*

This demo shows how Blobby recognizes the filters applied to the current page and uses them when answering questions. The walkthrough checks the active filters, asks about the data, and verifies that the resulting query respects the selected date range rather than querying all time.

[https://www.youtube.com/embed/Zihtd4iiCxM](https://www.youtube.com/embed/Zihtd4iiCxM)

## Visual Check Admin Settings for Dashboards

*`Ben Horne · Administration`*

This demo walks through an admin setting that controls when Blobby visually checks dashboard edits. It shows the options to run checks automatically, run them only when requested, or disable them entirely so the tool is no longer offered.

[https://www.youtube.com/embed/HpvcVl2af7I](https://www.youtube.com/embed/HpvcVl2af7I)

## Expanded MCP Read Tools

*`Ben Horne · MCP API`*

This demo introduces approximately 32 additional MCP read tools that expand the Omni functionality available through connected AI clients. It explains the focus on read operations and how the expanded toolset moves MCP access closer to the capabilities available through Omni’s interface and API.

[https://www.youtube.com/embed/3YZiFAbRlzo](https://www.youtube.com/embed/3YZiFAbRlzo)

## AI Hub Analytics Refresh

*`Arielle Strong · Analytics AI`*

This demo presents a refreshed AI Hub built with apps to make AI usage insights more focused and actionable. It discusses how session categorization, tagging, and sentiment analysis can support a clearer view of activity and help users identify useful next steps.

[https://www.youtube.com/embed/WA6YffUJ38c](https://www.youtube.com/embed/WA6YffUJ38c)

## Menu Consistency

*`Luke Bowerman · Dashboards`*

Full overhaul of menu changes that make navigation more consistent across apps, dashboards, and workbooks. It covers the placement of Develop actions, menu names that reflect the current tab type, matching right-click options, and a quick-action menu on the Develop page.

[https://www.youtube.com/embed/w3qWQon3dRE](https://www.youtube.com/embed/w3qWQon3dRE)

## Draft-Based Document Settings

*`Luke Bowerman · Dashboards`*

This demo shows how document settings are edited within a draft and applied to the published document when the draft is published. It walks through the read-only published view, editing a setting in a draft, and the separate confirmation flow for identifier changes that take effect immediately.

[https://www.youtube.com/embed/zdos\_XVqTR0](https://www.youtube.com/embed/zdos_XVqTR0)

## Autonomous Model Suggestions (AI Learning System)

*`Chris Swenson · Modeling AI Learning`*

A massive overhaul to our background agent that researches usage patterns and chat traces to identify improvements to a data model. It shows how the agent validates findings with queries, writes suggested changes to a branch, and summarizes its work, followed by a look at topic-based navigation in the suggestions interface.

[https://www.youtube.com/embed/RCXn9EnHMzw](https://www.youtube.com/embed/RCXn9EnHMzw)

## More Efficient Blobby

*`Dan Mosora · AI Token Management`*

This demo compares how context and cache improvements reduce the amount of work needed for a Blobby session. It then explains context collapsing, where large sections of an app are represented as expandable outlines so Blobby can load relevant details without reading the entire app upfront.

[https://www.youtube.com/embed/KfwK4tT6ZEc](https://www.youtube.com/embed/KfwK4tT6ZEc)

## GitHub MCP App (Read / Write)

*`Nathan Agrin · Apps AI MCP Github`*

This demo shows an Omni app using MCP connectors to interact with GitHub. It demonstrates writing back to the connected service by creating a ticket, then opening the result to confirm that the action completed successfully.

[https://www.youtube.com/embed/xey6e-rkFkM](https://www.youtube.com/embed/xey6e-rkFkM)

## App Performance Inspector (Query Speed Tool for Apps)

*`Nathan Agrin · Apps AI Performance`*

This demo explores the Performance tab in the app inspector and shows how it helps explain query execution times. It walks through the timing visualization, the map of queries on the page, a comparison without cached results, and options to inspect query details or open the underlying query.

[https://www.youtube.com/embed/vQ9vYJQfV6U](https://www.youtube.com/embed/vQ9vYJQfV6U)

## Omni Deep Links from Claude Apps (Plus Badges)

*`Nathan Agrin · AI MCP`*

This demo shows Claude creating a dashboard artifact with data retrieved through Omni’s MCP tools. It highlights an Omni source badge attached to the resulting artifact and demonstrates how clicking the badge takes the user back into Omni.

[https://www.youtube.com/embed/MqWgzhwXUrA](https://www.youtube.com/embed/MqWgzhwXUrA)

## Workbook Document Controls

*`Luke Bowerman · Dashboards Apps`*

This demo shows how document-level controls can work across workbook tabs while preserving selected filter values. It walks through control mapping and interactions in the published workbook, then explains how control sets can carry saved filter selections between surfaces within the same document.

[https://www.youtube.com/embed/HZgowbLajxU](https://www.youtube.com/embed/HZgowbLajxU)

## End Users Can Submit AI Modeling Suggestions

*`Chris Swenson · AI Modeling Learning`*

This demo shows how users can send ideas and model improvements to an admin suggestion inbox. It walks through submitting a plain-text idea and a suggestion with an implemented branch change, then explains how workbook-level modeling changes can enter the same review workflow.

[https://www.youtube.com/embed/Bo1Di62LVvE](https://www.youtube.com/embed/Bo1Di62LVvE)

## App Templates

*`Meredith Dodge · Apps AI`*

This demo shows how Blobby uses built-in templates to create dashboards and presentations with useful functionality already in place. The walkthrough covers dashboard filters, drilling, and loading states, followed by presentation navigation, full-screen presenting, and PDF and PowerPoint export controls.

[https://www.youtube.com/embed/zcwP1qTNWzU](https://www.youtube.com/embed/zcwP1qTNWzU)

## Python in Omni

*`Dan Spangenberger · Python AI Apps`*

This demo explores how Blobby uses Python for data analysis, chart generation, PCA-based apps, and joins between uploaded CSVs and Omni query results. It also presents a notebook-style interface that interleaves queries, visualizations, and Python steps, followed by questions about network access and available libraries.

[https://www.youtube.com/embed/CxXf8WsP\_GU](https://www.youtube.com/embed/CxXf8WsP_GU)

## Who Gets Which LLMs

*`Zef Lin · AI Administration`*

This demo shows how administrators can control which models are available to different users. It walks through configuring access for everyone, administrators, or selected audiences using organization roles, user attributes, connection roles, and user groups.

[https://www.youtube.com/embed/GoVQ30vfGe0](https://www.youtube.com/embed/GoVQ30vfGe0)

## Audience-Specific Default LLMs

*`Zef Lin · AI Administration`*

This demo shows how different audiences can receive different default models through configurable rules. It walks through examples for user groups and embedded users, then explains how rule order determines which default is selected when multiple audience conditions are evaluated.

[https://www.youtube.com/embed/EnDvhhg9jII](https://www.youtube.com/embed/EnDvhhg9jII)

## LLM Override for Routines

*`Eli Cohen · AI Routines Scheduler`*

This demo shows how to choose a model for an individual routine instead of always using the organization’s default. It creates a routine with a requested model, opens the model picker, and demonstrates how to change the selection or return to the default.

[https://www.youtube.com/embed/bsd2Igu-yUk](https://www.youtube.com/embed/bsd2Igu-yUk)

## AI Hub UX Tuning

*`Sarah Waterson · AI Apps`*

This demo shows how model selection moves into the AI Hub pages where it is needed rather than living in a separate location. It walks through the placement of model pickers for suggestions, evaluation runs, and prompt sets.

[https://www.youtube.com/embed/CjXqlTchcQU](https://www.youtube.com/embed/CjXqlTchcQU)

## Chart Rendering Performance

*`Richard Czechowski · Visualization Performance`*

This demo compares rendering times for a chart containing 50,000 rows before and after a set of performance optimizations. It shows an approximately 30% improvement and explains how avoiding unnecessary work, extra renders, and unused column decoding contributes to the result.

[https://www.youtube.com/embed/JoiADgueBRs](https://www.youtube.com/embed/JoiADgueBRs)

## Window Functions in the Model

*`Steven Talbot · Modeling Workbook Window Functions`*

This demo shows how window functions can be defined as measures in a model, with examples of running totals, lagged totals, partitioned calculations, and rolling windows. It demonstrates how the measures pull in required dimensions and how partitioning behaves when results are pivoted, followed by an explanation of how the definitions appear in a workbook.

[https://www.youtube.com/embed/GqMS-vQKV\_Y](https://www.youtube.com/embed/GqMS-vQKV_Y)

## Unlink Linked Tabs

*`Richard Czechowski · Apps Visualization`*

This demo shows how to unlink a tab from its source when a visualization needs its own independently editable query. It walks through the unlink action and confirmation, then shows how the resulting tab preserves its visualization and query-model changes while behaving like a normal tab.

[https://www.youtube.com/embed/z3a2pvcuXg0](https://www.youtube.com/embed/z3a2pvcuXg0)

## Live Embedded Pivot Exports to Excel

*`Conor O'Mara · Use Cases Apps`*

This demo shows how users can configure an embedded pivot, filter its data, and export the results into an editable spreadsheet pivot table. It walks through grouping metrics by region and vendor, downloading the matching rows, and opening the exported file to change the pivot’s dimensions in a spreadsheet application.

[https://www.youtube.com/embed/S-pZt1XbAVw](https://www.youtube.com/embed/S-pZt1XbAVw)