We've made several enhancements to our Data Apps feature, improving load times and security:

1. We have consolidated the way we preview and deploy Data Apps, resulting in faster load times when saving edits to your Data App. In some cases this could take over a minute, and now this takes just a couple of seconds.

2. Data Apps are now more secure so that you must be a NinjaCat user in order to open up a Data App (that has not been made public). See third point here...

3. If you are interested in sharing a Data App with a non-NinjaCat user, you still have the ability to do so, but we've just made it a more formal and secure method of sharing. There is now an option in the Configure tab of the Data App Builder to "Enable Public URL." This generates a public URL so that once the Data App is saved, this URL can be shared with others (so anyone with the link is able to access your Data App). You also have the option of later disabling this toggle. In doing so, the Data App results in an 'app not found' for any user that attempts to open the link.

Our webscraping tool that can optionally be enabled for Agents to scrape external websites (driven by Firecrawl) has been updated. Now when its used by an Agent, it has the ability to return all the formats that Firecrawl supports, if necessary (such as the raw HTML and image assets).

Also, we will now save any screenshots of the website that have been pulled back by the tool so that the Agent has the ability to visually analyze those screenshots (the website). Now, the user can work on things more effectively with the agent such as "tell me how to optimize our customer's website"

OpenAI has upgraded their GPT‑5 series with the release of GPT-5.2 Instant and GPT-5.2 Thinking. We now support both as model options in the Agent Builder to optionally choose for your Agents. To learn more, see OpenAI's support docs: https://openai.com/index/introducing-gpt-5-2/

Important note: the GPT-5 series (including, 5, 5.1, 5.2) all have larger context windows than older OpenAI models as well as Anthropic models. The GPT-5 series supports conversations up to 400k tokens, although in some cases may mean a slightly higher price. For a full view of our supported models and their cost + context window support, see our ReadMe doc: https://docs.ninjacat.io/docs/model-selection

We've added bulk import/export capabilities for Global Filters and Global Metrics. You can now download all your filters and metrics (including advertiser overrides) as a CSV, make changes in bulk, and upload them back in one operation. The import process validates your data, tracks progress in real-time, and provides detailed error reports if any issues occur. This is perfect for provisioning new configurations or updating metrics across multiple advertisers at scale.

Find the Export/Import buttons at Agency Settings → Bulk Management → Global Filters (or Global Metrics).

This feature is currently in Beta and available upon request. To request it, reach out to your Customer Success Manager or Support.

We've added a Network Health Checker to help you quickly diagnose connection issues. Click "Check Health" on any network to instantly verify your credentials are valid and all advertiser accounts are accessible. Results show immediately with detailed information about any failed connections, making it easy to identify and fix problems before they impact your reporting.

Access it from Settings → Networks → Edit any network → Check Health.

There is a new "Data Apps Access" permission toggle available at the user level. We have decoupled Data Apps access from our Data Cloud toggle so that users have more flexibility to permit access to certain users across our features. In order to View and Manage Data Apps, a user must have "Data Apps Access" toggle enabled for their User Role.

Resolve issues faster with Support by providing technical details with the click of a button. Navigate to the widget in question, right-click and select Copy Support Info. Paste that information into your support ticket. This will provide us everything we need to solve and troubleshoot.

We've successfully upgraded our Amazon DSP integration from API v2 to v3, ensuring continued compatibility with Amazon's latest reporting standards.

What Changed: As part of this upgrade, we've updated field names and deprecated certain metrics in accordance with Amazon's API migration. If you're currently using Amazon DSP in your templates or datasets, please note that any fields that have been deprecated by Amazon will now appear with "(deprecated)" in the template editor.

New Fields Added: We've added the following fields:

Creative Name Device Browser City Supply Source Name Placement Name Video event metrics: Video Start Events, Video Complete Events, Video First Quartile Events, Video Midpoint Events, Video Third Quartile Events, and Video Completion Rate

We now support Anthropic's latest model: Claude Opus 4.5, along with an extended 'Thinking' option - both available as model options in the Agent Builder.

Learn more about Claude Opus 4.5 from Anthropic: https://www.anthropic.com/news/claude-opus-4-5 and how it compares to its other models. If you are already using Claude Sonnet 4.5 with an Agent, it would be a worthy test to see how your output compares between that and the new Opus 4.5.

We've enhanced the reliability of AI Executive Summary widgets in your reports. Our improved system now includes smarter throttling and retry logic that ensures your AI-generated insights consistently appear in scheduled reports, even during high-volume periods. You should no longer see placeholder text in place of your AI analysis.