We can't find the internet
Attempting to reconnect
Something went wrong!
Attempting to reconnect
Give your team—and yourself—access to company data through the AI tools you choose. Set clear boundaries and see how that access is used.
Let your team ask questions of current company data in the AI tools they already use. Share a reviewed query with sales or support, choosing which information it returns and who can run it.
Use AI to explore the numbers behind a decision, with access to the databases you need. Review the queries and results, and manage your team's access as its needs change.
Coming soon · sites connected to live data
A sales team needs a clearer view of which accounts to follow up. With AI, they could turn that need into a shared workspace connected to company data—with access boundaries set by your organization.
We're bringing live data access to sites built with AI. Here's an example of the kind of tool your team could make.
Northstar
Sales workspace
YOUR WEEK AHEAD
A shared view of your accounts, open opportunities, and upcoming renewals.
Illustrative sales workspace with fictional accounts and sample figures. Live-data site connections are coming soon; this example is not connected to a database.
Explore what your team could build →Ask the people doing the work what would help. A view of upcoming renewals. An order tracker. A clearer picture of the week's results.
Use AI to shape the tool. Decide which data it needs and who should be able to use it. Review the result before sharing it.
Let colleagues try it, learn what helps, and refine it as the work changes. Your team decides what to improve next.
Available today · data access for your AI tools
Define the SQL and its parameters, preview the result, then publish it to the Catalog. Grant it to a person or group. Their AI can run that query without receiving general access to the underlying database.
Useful for recurring questions: revenue by month, outstanding orders, or a customer's support history. You decide which columns and rows the query returns.
Give people who need to explore a database permission to query it through Gentility. Read and write access are separate; database write protection and configured policies also apply.
A saved-query grant does not remove existing database permissions. Review both when deciding what someone's AI should be able to reach.
Draft and preview the SQL. Publish the reviewed version. Grant it to the people who need it, then inspect run history and maintain access as their work changes.
Saved queries currently require a direct PostgreSQL connection. Connector-routed databases and MySQL are not supported by saved queries yet.
See how it works →Your team may use different AI tools for different work. Gentility connects company data to compatible clients through MCP, including Claude, ChatGPT, and coding tools. Manage database permissions and saved-query grants in Gentility as you choose which clients to use.
Client features and provider data policies vary. Choosing a client does not by itself determine where all of your data is processed. See the data path.
Query results pass through Gentility to your AI client. Query records can include inputs and result samples. Your client and model provider also process the content you send them.
Review permissions, connection security, and sharing together. We explain the controls and their limits so you can decide what fits your organization.
Explore access and control →Start by giving people useful access to data. Gentility also offers Desk for working directly with coding agents, shared tasks and knowledge, and agents that respond to events or carry out recurring work.
You decide which responsibilities to delegate and which decisions need a person. Extend access as the work requires it, with explicit permissions for the databases an agent can use.
Connect a database, review a query, and choose who can use it.