Tools

Agents can use tools to interact with the external world, these can be APIs, databases, or other services, increasing the capabilities of your agent and extending its functionality beyond its own knowledge base and fixed environment.

Tool types

We provide a variety of different types of tools agents can use.

Type	Description
Web Search	Built-in. Search the web for up-to-date information not in the agent's knowledge.
Code Interpreter	Built-in. Run code and generate plots, useful for data analysis, visualization, or sandboxed execution.
Image Generation	Built-in. Generate images based on a prompt or description.
Document Library	Built-in. Search through documents uploaded to your Libraries, enabling RAG to answer questions based on specific information.
Function Calling	Custom local tools: functions defined in your environment that can be called by the agent. Execution happens locally.
Connectors	Register MCP servers as managed Connectors. Tools are discovered automatically and executed server-side.

Built-in tools

Built-in tools are ready out of the box. They can be called at any point, and all execution happens in our internal environment. They're also available for users to use directly via Conversations without creating an Agent first.

General usage

These built-in tools are only available when using the Agents and/or Conversations APIs.

To use them, specify the tool in the tools parameter when creating an Agent or calling the Conversations API.

tip

You can use multiple tools at the same time!

agent = client.beta.agents.create(
    model="<model>",
    name="<name_of_the_agent>",
    description="<description>",
    instructions="<instructions_or_system_prompt>",
    tools=[<list_of_tools>]
)

agent = client.beta.agents.create(
    model="<model>",
    name="<name_of_the_agent>",
    description="<description>",
    instructions="<instructions_or_system_prompt>",
    tools=[<list_of_tools>]
)