# Agent-Tool Interaction Via gRPC

An agent who wants to query the weather without having a weather tool/plugin, using Membase to find and utilize a similar tool:

1. **Identify the Need for Weather Information:** The agent recognizes the need to check the weather for a specific location but realizes that there is no pre-installed weather tool or plugin available.
2. **Search for Similar Tools:** Within Membase, the agent conducts a search for any tools or plugins that are similar to a weather tool. This involves querying the database for relevant keys or categories that might include weather-related services.
3. **Locate the Weather Plugin Service:** Once a suitable weather plugin is identified within the Membase database, the agent notes the service or server where the plugin is hosted. This information is crucial for the next steps.
4. **Load/Integrate the Weather Plugin Service:** The agent proceeds to load the service that hosts the weather plugin. This could involve establishing a connection to the server or initiating a session.
5. **Perform the Weather Query:** Now that the weather plugin is configured, the agent can perform the weather query. This typically involves sending a request to the plugin and receiving the current weather data for the specified location.
6. **Tool Discovery and Connection**

<figure><img src="/files/vYYMYeY0ZfWV0GqsGMbK" alt=""><figcaption><p><em>Search and connect to tool servers through LLM chat interface</em></p></figcaption></figure>

2. **Tool Usage**

<figure><img src="/files/tWAlhi6ajuBbB4eyWvD2" alt=""><figcaption><p><em>Interact with tools through LLM chat interface</em></p></figcaption></figure>


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://unibaseio.gitbook.io/unibase-docs/aip/agent/agent-tool-interaction-via-grpc.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.