August 8
What's New in QORE
We've been busy. Here's everything that landed recently - from a platform-wide reorganization to a long list of new integrations.
A New Shape for the Platform
We've restructured QORE into three clear spaces, each built around what you're actually trying to do:
Build - where you create apps, agents, and knowledge from the ground up.
Run - where those apps and agents actually run and get used (this space is still a work in progress - more coming soon).
Govern (Control Center) - where you keep watch over everything: governance, traceability, and observability into how your apps and agents are behaving.
Think of it as:
Build → Run → Govern
create it, use it, keep an eye on it. This is just the first step - these three spaces will keep evolving with every release.
Cost control features. We've started rolling out platform-wide cost controls, including auto-unpublishing inactive Dev/QA apps and cost visibility at build time for both apps and agents — more on this below.
Team-level access definition. You can now define access at the team level, making it easier to control who can see and do what across the platform.
Platform-level analytics. A new account-level analytics view shows overall usage across all your apps and agents, giving you a single place to see how the platform is being used.
App Builder
Import external JS apps. You can now import external JavaScript-based apps built on any other platform, including direct imports from GitLab and GitHub.
Scan & assessment report on import. When you import an external app, we now run a scanning and assessment report first, so you know what you're bringing in before it becomes part of your build.
Auto-generated data flow diagrams. Every app now gets an auto-generated data flow diagram (powered by Mermaid), so you can visualize how data moves through your app without drawing it yourself.
DB Explorer view. You've always had a managed DB — now you can actually see it. A new DB Explorer view lets you browse your database directly.
Auto testcase generation & AI-assisted fixes. Test cases are now auto-generated, test runs happen automatically, and if something breaks, AI can help fix it.
Security scanning with AI fixes. Apps are scanned for security issues, and AI can help you resolve what it finds.
Failure log visibility on publish. Failure logs are now visible right when an app publish fails — and AI can help you fix the issue on the spot.
Headless App Builder MCP. A headless App Builder MCP is now available, so you can build apps directly from Claude or any coding platform you prefer — no need to switch context.
Mobile backend support + store-ready packaging. Mobile app development now supports backend integration, and we've made improvements to get your app package much closer to Android Play Store–ready.
Agent conversation history in apps. When an agent is integrated into an app, its conversation history can now be shown within the app, depending on the use case.
Auto-unpublish on Dev/QA after 5 hours. In line with our ongoing cost-control efforts, apps on Dev and QA environments will now auto-unpublish after 5 hours of inactivity.
In-app and unified app analytics. In-app analytics is now available, alongside a unified app analytics view in the Control Center.
App build-time cost visibility. App build-time cost visibility has been added, so you know what you're spending before you even publish.
Agent Builder
PII tokenization in agent guardrails. PII tokenization is now part of agent guardrails. Based on your configured PII service provider, sensitive information gets automatically redacted — both at runtime and in the trace logs.
Agent build-time cost visibility. Agent build-time cost visibility is now available.
Cost control for runaway/looping agents. New cost control mechanisms kick in if an agent starts looping or spiraling out of control, keeping runaway costs in check.
Online evals for deployed agents. The Control Center now includes online evals, so you can test your agents even after they've been deployed.
Agent analytics with action traceback. Agent analytics now shows a complete traceback of agent actions — bringing the observability we previously relied on OPIK for, natively into the platform.
Improved knowledge pipeline. The knowledge pipeline has been improved — and there's more on the way.
New Integrations
We've added a wave of new integrations to help QORE connect with the tools you already use. Most have been fully tested with dev credentials; a few are pending real-world credentials for full validation:
Productivity and Collaboration
Microsoft Teams
Calendly
Fathom
Notion
Airtable
Discord
Telegram
Fireflies.ai
Microsoft Excel
Gmail (G-Suite)
Jira
CRM and Marketing
Google Places
Apollo.io
Microsoft Dynamics 365
PandaDoc
Zendesk
Shopify
Mailchimp
Banking and Finance
Recurly
Middesk
Xerox
AI and Utilities
Supabase
Firecrawl
Compliance and Analytics
Docusign: Document Management, OCR, eSign
Google Analytics
As always, this is just the beginning—the Build, Run, and Govern spaces will keep growing with each release.
Thanks for building with us!
Last updated