What are slot resolvers?

Slot resolvers convert natural language into data types. The way that users will reference your business objects will seldom match how they're stored in business systems.

For example...

• "Tomorrow's Standup" is a GoogleCalendarEvent with ID = 4s567d8s908f87654sa678ds
• "Jamie" is a User with ID = 9e107d9d-372b-4ac9-b4e9-0fbccd3029ab

We purpose-built Slot Resolvers to solve this problem. Plugins built using slot resolvers will perform substantially better when deployed to production.

How do they work?

Plugin Selection

First, your Moveworks AI Assistant needs to select a plugin that uses slots resolvers. Slot resolvers are attached to the slots on your plugin.

If you've selected a resolver strategy for your slot, or if your slot uses an object data type, then the slot resolver begins.

Data Type Retrieval

Then, the AI agent retrieves possible values that might satisfy the data type constraint of your slot. These are retrieved using the resolver strategy.

Disambiguation

Then, when the AI agent finds multiple possible matches for a business object, it presents them (with citations) so the user can pick the right one.

State Management

Throughout the process, the AI agent keeps track of the possible values & the selected value using a symbolic memory architecture. This is part of our Agentic Reasoning Engine!

You can be confident that LLMs will not corrupt or get confused by IDs in the conversation.

Action Usage

Once slot resolvers collect the necessary business objects, they will be passed to your plugin as either a list or a single instance of the slot's data type.

See our slots documentation for more on how to use these in your plugin.