---
title: "Agent tools · the built-in capabilities agents call · Agent Fieldbook"
url: https://agentfieldbook.org/software/tools/
description: "The canonical built-in tools an AI agent calls: web search, code execution, computer use, file I/O, retrieval, memory, MCP connectors, image generation, function calling and tool search. A source-graded taxonomy showing how OpenAI, Anthropic and Google each name every capability, linked to official provider docs."
section: "Stack · Agent tools"
source: Agent Fieldbook — generated from the published page
---

# The agent toolbox

A model on its own only writes text. An agent is a model given tools: the capabilities it invokes to search, run code, drive a computer, read files and remember. Here is the common core, capability by capability, sourced to each provider's docs.

13 built-in tool types every serious agent platform exposes, grouped by what they do. 7 are near-universal (OpenAI, Anthropic and Google all ship them); the rest are emerging, shipped by one or two providers so far. Each is the capability itself, not a vendor product: the identifiers show how OpenAI, Anthropic and Google name it, and every row links to a provider's official tool documentation.

## Information & retrieval · 3

Run live web queries and ground the answer in current sources with citations, rather than relying on the model's training cut-off.

- [↗ provider docs](https://developers.openai.com/api/docs/guides/tools-web-search)

Retrieve the full content of a specific URL the agent already holds, as opposed to searching the open web for one.

- [↗ provider docs](https://platform.claude.com/docs/en/agents-and-tools/tool-use/web-fetch-tool)

Retrieve the relevant chunks from a developer's uploaded documents or vector store into the model's context (retrieval-augmented generation).

- [↗ provider docs](https://developers.openai.com/api/docs/guides/tools-file-search)

## Execution · 3

Write and run code, typically Python, in a hosted sandbox for computation, data analysis and mathematics.

- [↗ provider docs](https://developers.openai.com/api/docs/guides/tools-code-interpreter)

Execute arbitrary shell commands in a runtime environment, local or hosted, giving the agent a terminal.

- [↗ provider docs](https://platform.claude.com/docs/en/agents-and-tools/tool-use/bash-tool)

Generate or edit images from a text prompt as a callable step inside the agent loop.

- [↗ provider docs](https://developers.openai.com/api/docs/guides/tools)

## File I/O · 1

View, create and edit files on disk via structured commands such as view, create, str_replace and insert.

- [↗ provider docs](https://platform.claude.com/docs/en/agents-and-tools/tool-use/text-editor-tool)

## Control · 1

Drive a graphical interface by taking screenshots and issuing mouse and keyboard actions to control a computer or browser.

- [↗ provider docs](https://platform.claude.com/docs/en/agents-and-tools/tool-use/computer-use-tool)

## Memory · 1

Persist and recall information across turns and sessions through a dedicated memory store the agent can read and write.

- [↗ provider docs](https://platform.claude.com/docs/en/agents-and-tools/tool-use/memory-tool)

## Integration & extensibility · 4

Connect the model to third-party tools and data exposed by a remote Model Context Protocol server, without hand-wiring each integration.

- [↗ provider docs](https://platform.claude.com/docs/en/agents-and-tools/mcp-connector)

Call developer-defined functions with structured JSON arguments; the universal substrate every other built-in tool is built on.

- [↗ provider docs](https://ai.google.dev/gemini-api/docs/function-calling)

Query structured databases and warehouses; delivered through MCP database connectors or by running SQL inside code execution rather than a first-party built-in.

- [↗ provider docs](https://developers.openai.com/api/docs/guides/tools-connectors-mcp)

Dynamically discover and lazily load the relevant tool definitions at runtime, saving context when many tools are available.

- [↗ provider docs](https://platform.claude.com/docs/en/agents-and-tools/tool-use/tool-search-tool)

## Field notes

The built-in tools every agent calls: search, code, computer use, files, memory and MCP

Thirteen built-in capabilities, the things an agent can actually do once it has a model: search and fetch the web, run code, use a shell, read and write files, drive a computer, retrieve from files, hold memory, call MCP connectors, generate images, call functions, query data and search its own tools.

These are the primitives every vendor now ships in some form, which is why the interesting question is no longer whether an agent has them but what it is permitted to do with them. Each one is also an entry point: web fetch and file read are how untrusted content reaches the model in the first place.

- [Brandon Chaplin](https://www.linkedin.com/in/brandon-chaplin-digital-marketing-strategist)

## Common questions

### What built-in tools do AI agents have?

A common core recurs across providers: web search, web fetch, code execution, a shell, file read and write, computer use, retrieval, memory, image generation and MCP connectors. Each vendor names them differently but the capabilities line up. Function calling is the mechanism underneath all of them.

### What is the difference between a built-in tool and function calling?

Function calling is the mechanism: the model emits a structured request naming a tool and its arguments. A built-in tool is a capability the provider ships and often runs on its own servers, such as web search or code execution, exposed through that same mechanism. Custom tools you write use the same substrate. [(OpenAI)](https://platform.openai.com/docs/guides/function-calling)

### Which tools do all the major AI platforms provide?

Web search, code execution, computer use, file search and function calling are shipped by OpenAI, Anthropic and Google alike. MCP connectors are close behind. Memory, standalone web fetch and tool search are still uneven, offered by only one or two providers.

### How do agents get tools their model provider does not ship?

Through MCP. An agent connects to a server that exposes third-party tools, so database access or a CRM integration works without the model provider building each one. Where no server exists, running code inside the sandbox fills the gap. [(Model Context Protocol)](https://modelcontextprotocol.io/)

### Is an agent tool the same as an agent framework?

No. Tools are the capabilities an agent invokes. A framework is the library you build the agent on, deciding when and in what order those tools get called. LangGraph and the OpenAI Agents SDK are frameworks; web search is a tool.

## Next in the learning path

- [MCP The protocol that extends an agent with third-party tools](https://agentfieldbook.org/software/mcp/)

- [Frameworks & Tools The libraries that orchestrate these tool calls](https://agentfieldbook.org/software/frameworks/)

- [Protocols How agents call tools and talk to each other](https://agentfieldbook.org/software/protocols/)

- [Harnesses The agent SDKs that wire these tools in](https://agentfieldbook.org/software/frameworks/)

- [Building blocks The architectural parts inside an agent](https://agentfieldbook.org/design/building_blocks/)

## Entries

_13 entries listed on this page._

| name | description | subjectOf |
| --- | --- | --- |
| Web search | Run live web queries and ground the answer in current sources with citations, rather than relying on the model's training cut-off. | https://developers.openai.com/api/docs/guides/tools-web-search |
| Web fetch | Retrieve the full content of a specific URL the agent already holds, as opposed to searching the open web for one. | https://platform.claude.com/docs/en/agents-and-tools/tool-use/web-fetch-tool |
| Code execution | Write and run code, typically Python, in a hosted sandbox for computation, data analysis and mathematics. | https://developers.openai.com/api/docs/guides/tools-code-interpreter |
| Shell / bash | Execute arbitrary shell commands in a runtime environment, local or hosted, giving the agent a terminal. | https://platform.claude.com/docs/en/agents-and-tools/tool-use/bash-tool |
| File read / write | View, create and edit files on disk via structured commands such as view, create, str_replace and insert. | https://platform.claude.com/docs/en/agents-and-tools/tool-use/text-editor-tool |
| Computer use | Drive a graphical interface by taking screenshots and issuing mouse and keyboard actions to control a computer or browser. | https://platform.claude.com/docs/en/agents-and-tools/tool-use/computer-use-tool |
| File search / retrieval | Retrieve the relevant chunks from a developer's uploaded documents or vector store into the model's context (retrieval-augmented generation). | https://developers.openai.com/api/docs/guides/tools-file-search |
| Memory | Persist and recall information across turns and sessions through a dedicated memory store the agent can read and write. | https://platform.claude.com/docs/en/agents-and-tools/tool-use/memory-tool |
| MCP connector | Connect the model to third-party tools and data exposed by a remote Model Context Protocol server, without hand-wiring each integration. | https://platform.claude.com/docs/en/agents-and-tools/mcp-connector |
| Image generation | Generate or edit images from a text prompt as a callable step inside the agent loop. | https://developers.openai.com/api/docs/guides/tools |
| Function calling | Call developer-defined functions with structured JSON arguments; the universal substrate every other built-in tool is built on. | https://ai.google.dev/gemini-api/docs/function-calling |
| Data / SQL access | Query structured databases and warehouses; delivered through MCP database connectors or by running SQL inside code execution rather than a first-party built-in. | https://developers.openai.com/api/docs/guides/tools-connectors-mcp |
| Tool search | Dynamically discover and lazily load the relevant tool definitions at runtime, saving context when many tools are available. | https://platform.claude.com/docs/en/agents-and-tools/tool-use/tool-search-tool |