***

title: Slot Resolvers
position: 2
excerpt: Resolving natural language into business objects
deprecated: false
hidden: false
---------------------

For clean Markdown of any page, append .md to the page URL. For a complete documentation index, see https://help.moveworks.com/agent-studio/core-concepts/agentic-automation-engine/llms.txt. For full documentation content, see https://help.moveworks.com/agent-studio/core-concepts/agentic-automation-engine/llms-full.txt.

# What are slot resolvers?

LLMs hallucinate. When a user says "Jamie" and there are 12 Jamies in your directory, the LLM picks the most probable one — which may not be the right one. One wrong ID passed to ServiceNow means the wrong person loses access. Slot resolvers exist to make this impossible.

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.

![](https://files.readme.io/24b4e8795e72b5c671a9f536ecd7e4ec1cbe92559915381fdcefd98135e71658-CleanShot_2024-10-24_at_18.18.212x.png)

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](/agent-studio/core-concepts/conversational-plugins) built using slot resolvers will perform substantially better when deployed to production.**

# How do they work?

## Candidate Retrieval

Then, the AI agent retrieves possible values that might satisfy the constraints of your slot. These are retrieved using the [resolver strategy](/agent-studio/conversation-process/resolver-strategies).

![](https://files.readme.io/3d741695b6e7dbf01a267e14f57d29e9d3d0c0303ece9a23d9c4ba7e6fe2cd3a-CleanShot_2025-02-05_at_12.17.08.png)

<br />

## Disambiguation

Then, when the AI agent finds multiple possible matches, it presents them (with [citations](/agent-studio/core-concepts/citations-1)) so the user can pick the right one.

![](https://files.readme.io/b760cb0254cda14f5fadf33ea042da84213e6e7dad75264af606eff348d1d096-CleanShot_2024-10-24_at_18.29.322x.png)

## Stable Memory & State Management

Slot Resolvers use a **symbolic working memory architecture** to keep track of both

1. the list of possible values possible values &
2. the selected value

As a result you can be confident the AI agent won't accidentally change IDs or make up new ones when providing them to your actions.

## Learn More

* [Resolver Strategies](/agent-studio/conversation-process/resolver-strategies) — full configuration reference for static and dynamic methods, output mapping, and context passing
* [Data Types](/agent-studio/core-concepts/data-types) — how to create custom data types and attach reusable resolver strategies
* [Slots](/agent-studio/conversation-process/slots) — how slots use resolver strategies to collect user input