This month's updates center on the quality of your agent's answers and the quality of what feeds them: Thinking Mode is now the default way new agents answer, content ingestion understands more of the media you add, and the app got faster in the places you and your seekers visit every day. Here's what landed.
Improved Answer Quality with Thinking Mode
We've had great success with Thinking Mode (formerly called "Tool Use"), which lets your agent take a moment to think and make multiple knowledge base searches — optionally searching the web as well — before responding. Answers to sophisticated questions come back more thorough and better sourced.
Thinking Mode is now the default for new agents. If your agent was created earlier, it's worth a moment to switch it on: go to Agents, select your agent, and enable it in the Thinking Mode section of the settings.

Learn more about agent settings
Better Transcription and Media Handling
Experts pull in media of many types and from a wide variety of web sites, and we're continually improving our ability to ingest relevant content and media from the links you add and the files you upload. This month:
- Embedded media on web pages is detected and extracted more reliably, so a page that hosts a video or podcast player yields the media itself, not just the text around it.
- Audio transcription is faster and more accurate, with improved speaker attribution so multi-speaker recordings read the way the conversation actually went.
- More RSS and feed types are recognized, so more of the feeds you add just work.
Learn more about adding content to your Knowledge Hub
Performance Improvements
We also spent time this month on speed in key areas of the product experience: expert page rendering, account creation, sign in, and content ingestion are all noticeably faster.
As always, this release includes many small improvements and bug fixes! We'd love to hear how these updates land for you — share your feedback with our CEO Scott's agent at answersfrom.me/scott.
Cheers,
The Answers From Me Team
Was this article helpful?
That’s Great!
Thank you for your feedback
Sorry! We couldn't be helpful
Thank you for your feedback
Feedback sent
We appreciate your effort and will try to fix the article