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.
We've added five new fonts to the Template Builder: Tungsten, Niveau (Niveau Grotesk), DM Sans, Inter, and Abril Fatface. These professional typefaces give you even more options to match your brand guidelines and create polished, on-brand reports for your clients.
OpenAI has upgraded their GPT‑5 series with the release of GPT-5.1 Instant and GPT-5.1 Thinking. We now support both as model options in the Agent Builder to optionally choose for your Agents. These models are described by OpenAI as the following, but learn more from their support docs here: https://openai.com/index/gpt-5-1/
GPT‑5.1 Instant: our most-used model, now warmer, more intelligent, and better at following your instructions. GPT‑5.1 Thinking: our advanced reasoning model, now easier to understand and faster on simple tasks, more persistent on complex ones.
Try testing your Agents out w/ GPT-5.1- we'd love to hear your feedback.
Agent Builder Bob..... Bob the BUILDER.... can he fix it? YES HE CAN. BOB now has the ability to test the agent that he is building (with / for the user) when the new "Test" mode option is enabled for that conversation (and then evaluate, and fix / adjust).
When Test mode is enabled, Bob has additional instructions to test the Agent after he has gone through all the building steps with the user, including testing any tools, custom actions, subagents, etc that have been set up for that Agent. He'll also give the user an evaluation of the Agent's output against what its set of instructions was. We're still working on ironing this part out, so over the next week or so, you may see the behavior of his testing and evaluation change. You may then instruct him to make necessary adjustments to the agent based on his evaluation. Test out the 'Test' mode and provide us with any feedback!
We've upgraded our LinkedIn API connection to 202510. This upgrade ensures our integration delivers ongoing support, and maintains optimal performance for your campaigns.
We've added dimension filtering to your report configurations! You can now filter your entire report by Campaign Name, Device Type, Network, Account, Data Source, and Custom Dimensions 1-10. This makes it easy to create focused reports for specific accounts or campaigns without having to manually configure individual widgets.
We're excited to announce that you can now restore deleted accounts directly from the NinjaCat platform!
This new feature gives you the flexibility to recover accounts that were previously deleted, eliminating the need for support tickets or manual intervention.
On the Accounts List view, you can filter by Status for Deleted accounts and restore your accounts from the list view.
We have recently introduced Version History to the Agent Builder! This gives users the ability to revert their Agent's set of Instructions to a past state, including the model assigned. When another version is selected in the Agent's Version History, the Instructions & Model will change while the rest of the settings of your Agent will remain the same as your current version (i.e. Custom Actions, Datasets, Tools, etc.).
We have added several fields to our StackAdapt Integration:
- Ad Name
- Demographic Segment
- City
- Zip / Postal Code
We have added several fields to our TikTok Integration:
- Ad Secondary Status
- Advertiser Currency
- Advertiser Name
- Advertiser Time Zone
- Campaign Bid Type
- Campaign Optimization Goal
- Campaign Secondary Status