The latest news, updates, and stories from Retool.

Do your AI agents need an org chart?
89% of IT leaders have AI agents in production, but only 27% have one team that owns them all. New survey data on AI agent governance and ownership. URL Slug: /blog/ai-agent-org-chart

How our AI support agent ran 851 investigations through Retool’s MCP server
Retool's support team runs a shared AI agent through Retool's MCP server. See how existing permissions, audit logs and grading powered 851 investigations.

Most of your AI agents don’t deserve to exist
AI ROI is real but concentrated. Retool CEO David Hsu joined a panel discussing why 90% of tokens are net negative, and how to find the 10% that pays off.
How our AI support agent ran 851 investigations through Retool’s MCP server
Retool's support team runs a shared AI agent through Retool's MCP server. See how existing permissions, audit logs and grading powered 851 investigations.

Give scheduled workflows the access they need with service users
Service users let scheduled and webhook workflows run with a permission group's access instead of an admin's, and they work with Data Access Enforcement.

How I use Retool as an MCP gateway to production data (as a software engineer)
How a Retool software engineer uses the Retool MCP server as an MCP gateway to production data: read-only queries, existing permissions, and full audit logs.

Do your AI agents need an org chart?
89% of IT leaders have AI agents in production, but only 27% have one team that owns them all. New survey data on AI agent governance and ownership. URL Slug: /blog/ai-agent-org-chart

How Retool governs AI-built apps that touch production data
How Retool keeps AI-built apps governed: access set once, credentials out of app code, unsafe changes blocked, and every query logged. Plus 5 questions to ask.

Put your cloud commitment to work: Retool is now on Google Cloud Marketplace
Retool is now on Google Cloud Marketplace. Put committed cloud spend toward Retool, skip new-vendor onboarding, and keep one invoice.

Data Access Enforcement: Set data permissions once and enforce access on every query
Set table-, column-, and row-level access policies once on a Postgres resource, and Retool enforces them on every query, from every app, workflow, and AI agent.

Retool launches the first centralized agent harness
Give every employee an AI agent in Slack, Teams, and Google Chat. Retool's centralized enterprise agent harness controls their tools, data, and models. Join the waitlist.

Retool and MongoDB partner to make it easier to build securely on Atlas
Learn how Retool and MongoDB are partnering to bring automated, secure credentials to MongoDB Atlas, enabling teams to build production-ready apps faster without compromising security.

Most of your AI agents don’t deserve to exist
AI ROI is real but concentrated. Retool CEO David Hsu joined a panel discussing why 90% of tokens are net negative, and how to find the 10% that pays off.

Build and deploy apps with Retool’s CLI
Choose the right tool for the job: how Retool's updated CLI compares to the MCP server, and when to build from your terminal instead of Claude Chat.

Retool MCP vs. CLI: When to use each building surface
Not sure whether to reach for Retool's MCP or CLI? MCP is built for reading, admin tasks, and chat-only surfaces. CLI is built for code-heavy app development.
