Skip to main content

Claude Agent SDK for Python

Python SDK for Claude Agent. See the Claude Agent SDK documentation for more information.

Installation

pip install claude-agent-sdk

Prerequisites:

  • Python 3.10+

Note: The Claude Code CLI is automatically bundled with the package - no separate installation required! The SDK will use the bundled CLI by default. If you prefer to use a system-wide installation or a specific version, you can:

  • Install Claude Code separately: curl -fsSL https://claude.ai/install.sh | bash
  • Specify a custom path: ClaudeAgentOptions(cli_path="/path/to/claude")

Quick Start

import anyio
from claude_agent_sdk import query

async def main():
    async for message in query(prompt="What is 2 + 2?"):
        print(message)

anyio.run(main)

Basic Usage: query()

query() is an async function for querying Claude Code. It returns an AsyncIterator of response messages. See src/claude_agent_sdk/query.py.

from claude_agent_sdk import query, ClaudeAgentOptions, AssistantMessage, TextBlock

# Simple query
async for message in query(prompt="Hello Claude"):
    if isinstance(message, AssistantMessage):
        for block in message.content:
            if isinstance(block, TextBlock):
                print(block.text)

# With options
options = ClaudeAgentOptions(
    system_prompt="You are a helpful assistant",
    max_turns=1
)

async for message in query(prompt="Tell me a joke", options=options):
    print(message)

Using Tools

By default, Claude has access to the full Claude Code toolset (Read, Write, Edit, Bash, and others). allowed_tools is a permission allowlist: listed tools are auto-approved, and unlisted tools fall through to permission_mode and can_use_tool for a decision. It does not remove tools from Claude's toolset. To block specific tools, use disallowed_tools. See the permissions guide for the full evaluation order.

options = ClaudeAgentOptions(
    allowed_tools=["Read", "Write", "Bash"],  # auto-approve these tools
    permission_mode='acceptEdits'  # auto-accept file edits
)

async for message in query(
    prompt="Create a hello.py file",
    options=options
):
    # Process tool use and results
    pass

Working Directory

from pathlib import Path

options = ClaudeAgentOptions(
    cwd="/path/to/project"  # or Path("/path/to/project")
)

ClaudeSDKClient

ClaudeSDKClient supports bidirectional, interactive conversations with Claude Code. See src/claude_agent_sdk/client.py.

Unlike query(), ClaudeSDKClient additionally enables custom tools and hooks, both of which can be defined as Python functions.

Custom Tools (as In-Process SDK MCP Servers)

A custom tool is a Python function that you can offer to Claude, for Claude to invoke as needed.

Custom tools are implemented in-process MCP servers that run directly within your Python application, eliminating the need for separate processes that regular MCP servers require.

For an end-to-end example, see MCP Calculator.

Creating a Simple Tool

from claude_agent_sdk import tool, create_sdk_mcp_server, ClaudeAgentOptions, ClaudeSDKClient

# Define a tool using the @tool decorator
@tool("greet", "Greet a user", {"name": str})
async def greet_user(args):
    return {
        "content": [
            {"type": "text", "text": f"Hello, {args['name']}!"}
        ]
    }

# Create an SDK MCP server
server = create_sdk_mcp_server(
    name="my-tools",
    version="1.0.0",
    tools=[greet_user]
)

# Use it with Claude. allowed_tools pre-approves the tool so it runs
# without a permission prompt; it does not control tool availability.
options = ClaudeAgentOptions(
    mcp_servers={"tools": server},
    allowed_tools=["mcp__tools__greet"]
)

async with ClaudeSDKClient(options=options) as client:
    await client.query("Greet Alice")

    # Extract and print response
    async for msg in client.receive_response():
        print(msg)

Benefits Over External MCP Servers

  • No subprocess management - Runs in the same process as your application
  • Better performance - No IPC overhead for tool calls
  • Simpler deployment - Single Python process instead of multiple
  • Easier debugging - All code runs in the same process
  • Type safety - Direct Python function calls with type hints

Migration from External Servers

# BEFORE: External MCP server (separate process)
options = ClaudeAgentOptions(
    mcp_servers={
        "calculator": {
            "type": "stdio",
            "command": "python",
            "args": ["-m", "calculator_server"]
        }
    }
)

# AFTER: SDK MCP server (in-process)
from my_tools import add, subtract  # Your tool functions

calculator = create_sdk_mcp_server(
    name="calculator",
    tools=[add, subtract]
)

options = ClaudeAgentOptions(
    mcp_servers={"calculator": calculator}
)

Mixed Server Support

You can use both SDK and external MCP servers together:

options = ClaudeAgentOptions(
    mcp_servers={
        "internal": sdk_server,      # In-process SDK server
        "external": {                # External subprocess server
            "type": "stdio",
            "command": "external-server"
        }
    }
)

Hooks

A hook is a Python function that the Claude Code application (not Claude) invokes at specific points of the Claude agent loop. Hooks can provide deterministic processing and automated feedback for Claude. Read more in Intercept and control agent behavior with hooks.

For more examples, see examples/hooks.py.

Example

from claude_agent_sdk import ClaudeAgentOptions, ClaudeSDKClient, HookMatcher

async def check_bash_command(input_data, tool_use_id, context):
    tool_name = input_data["tool_name"]
    tool_input = input_data["tool_input"]
    if tool_name != "Bash":
        return {}
    command = tool_input.get("command", "")
    block_patterns = ["foo.sh"]
    for pattern in block_patterns:
        if pattern in command:
            return {
                "hookSpecificOutput": {
                    "hookEventName": "PreToolUse",
                    "permissionDecision": "deny",
                    "permissionDecisionReason": f"Command contains invalid pattern: {pattern}",
                }
            }
    return {}

options = ClaudeAgentOptions(
    allowed_tools=["Bash"],
    hooks={
        "PreToolUse": [
            HookMatcher(matcher="Bash", hooks=[check_bash_command]),
        ],
    }
)

async with ClaudeSDKClient(options=options) as client:
    # Test 1: Command with forbidden pattern (will be blocked)
    await client.query("Run the bash command: ./foo.sh --help")
    async for msg in client.receive_response():
        print(msg)

    print("\n" + "=" * 50 + "\n")

    # Test 2: Safe command that should work
    await client.query("Run the bash command: echo 'Hello from hooks example!'")
    async for msg in client.receive_response():
        print(msg)

Types

See src/claude_agent_sdk/types.py for complete type definitions:

  • ClaudeAgentOptions - Configuration options
  • AssistantMessage, UserMessage, SystemMessage, ResultMessage - Message types
  • TextBlock, ToolUseBlock, ToolResultBlock - Content blocks

Error Handling

from claude_agent_sdk import (
    ClaudeSDKError,      # Base error
    CLINotFoundError,    # Claude Code not installed
    CLIConnectionError,  # Connection issues
    ProcessError,        # Process failed
    ResultError,         # Run ended with an error result (subclass of ProcessError)
    CLIJSONDecodeError,  # JSON parsing issues
)

try:
    async for message in query(prompt="Hello"):
        pass
except CLINotFoundError:
    print("Please install Claude Code")
except ResultError as e:
    # The CLI reported a terminal error result (also yielded as the final
    # ResultMessage) and exited. Structured fields are on the exception.
    print(f"Run failed: {e.subtype=} {e.terminal_reason=} {e.api_error_status=}")
    print(e.result or e.errors)
except ProcessError as e:
    print(f"Process failed with exit code: {e.exit_code}")
except CLIJSONDecodeError as e:
    print(f"Failed to parse response: {e}")

See src/claude_agent_sdk/_errors.py for all error types.

Available Tools

See the Claude Code documentation for a complete list of available tools.

Examples

See examples/quick_start.py for a complete working example.

See examples/streaming_mode.py for comprehensive examples involving ClaudeSDKClient. You can even run interactive examples in IPython from examples/streaming_mode_ipython.py.

Migrating from Claude Code SDK

If you're upgrading from the Claude Code SDK (versions < 0.1.0), please see the CHANGELOG.md for details on breaking changes and new features, including:

  • ClaudeCodeOptionsClaudeAgentOptions rename
  • Merged system prompt configuration
  • Settings isolation and explicit control
  • New programmatic subagents and session forking features

Development

If you're contributing to this project, run the initial setup script to install git hooks:

./scripts/initial-setup.sh

This installs a pre-push hook that runs lint checks before pushing, matching the CI workflow. To skip the hook temporarily, use git push --no-verify.

Building Wheels Locally

To build wheels with the bundled Claude Code CLI:

# Install build dependencies
pip install build twine

# Build wheel with bundled CLI
python scripts/build_wheel.py

# Build with specific version
python scripts/build_wheel.py --version 0.1.4

# Build with specific CLI version
python scripts/build_wheel.py --cli-version 2.0.0

# Clean bundled CLI after building
python scripts/build_wheel.py --clean

# Skip CLI download (use existing)
python scripts/build_wheel.py --skip-download

The build script:

  1. Downloads Claude Code CLI for your platform
  2. Bundles it in the wheel
  3. Builds both wheel and source distribution
  4. Checks the package with twine

See python scripts/build_wheel.py --help for all options.

Release Workflow

The package is published to PyPI via the GitHub Actions workflow in .github/workflows/publish.yml. To create a new release:

  1. Trigger the workflow manually from the Actions tab with two inputs:

    • version: The package version to publish (e.g., 0.1.5)
    • claude_code_version: The Claude Code CLI version to bundle (e.g., 2.0.0 or latest)
  2. The workflow will:

    • Build platform-specific wheels for macOS, Linux, and Windows
    • Bundle the specified Claude Code CLI version in each wheel
    • Build a source distribution
    • Publish all artifacts to PyPI
    • Create a release branch with version updates
    • Open a PR to main with:
      • Updated pyproject.toml version
      • Updated src/claude_agent_sdk/_version.py
      • Updated src/claude_agent_sdk/_cli_version.py with bundled CLI version
      • Auto-generated CHANGELOG.md entry
  3. Review and merge the release PR to update main with the new version information

The workflow tracks both the package version and the bundled CLI version separately, allowing you to release a new package version with an updated CLI without code changes.

License and terms

Use of this SDK is governed by Anthropic's Commercial Terms of Service, including when you use it to power products and services that you make available to your own customers and end users, except to the extent a specific component or dependency is covered by a different license as indicated in that component's LICENSE file.

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

claude_agent_sdk-0.2.148.tar.gz (344.7 kB view details)

Uploaded Source

Built Distributions

If you're not sure about the file name format, learn more about wheel file names.

claude_agent_sdk-0.2.148-py3-none-win_amd64.whl (96.9 MB view details)

Uploaded Python 3Windows x86-64

claude_agent_sdk-0.2.148-py3-none-manylinux_2_17_x86_64.whl (94.2 MB view details)

Uploaded Python 3manylinux: glibc 2.17+ x86-64

claude_agent_sdk-0.2.148-py3-none-manylinux_2_17_aarch64.whl (94.0 MB view details)

Uploaded Python 3manylinux: glibc 2.17+ ARM64

claude_agent_sdk-0.2.148-py3-none-macosx_11_0_x86_64.whl (88.3 MB view details)

Uploaded Python 3macOS 11.0+ x86-64

claude_agent_sdk-0.2.148-py3-none-macosx_11_0_arm64.whl (83.9 MB view details)

Uploaded Python 3macOS 11.0+ ARM64

File details

Details for the file claude_agent_sdk-0.2.148.tar.gz.

File metadata

  • Download URL: claude_agent_sdk-0.2.148.tar.gz
  • Upload date:
  • Size: 344.7 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/7.0.0 CPython/3.12.14

File hashes

Hashes for claude_agent_sdk-0.2.148.tar.gz
Algorithm Hash digest
SHA256 45c9972fa72dc6006745239c6838cc3cea7a9b42d6927c89f7bdb6996a1983b0
MD5 19dfaa5788ca03b16ee26cbcb436f208
BLAKE2b-256 7ef4fb7f81b31e44f69af9f53b93c32e2047f3ca15cd6311cfe72bfdf420f1f5

See more details on using hashes here.

File details

Details for the file claude_agent_sdk-0.2.148-py3-none-win_amd64.whl.

File metadata

File hashes

Hashes for claude_agent_sdk-0.2.148-py3-none-win_amd64.whl
Algorithm Hash digest
SHA256 f155252940fd77f1bcc707836924ca3a9691d19bbc464914ae2b7ccf17f0410f
MD5 deacc7653f13dc047f5797d60675c26a
BLAKE2b-256 c3ea86ccd7a34446c37174e53f03c389b25f168c7f4fd4f00f3e130baae9c902

See more details on using hashes here.

File details

Details for the file claude_agent_sdk-0.2.148-py3-none-manylinux_2_17_x86_64.whl.

File metadata

File hashes

Hashes for claude_agent_sdk-0.2.148-py3-none-manylinux_2_17_x86_64.whl
Algorithm Hash digest
SHA256 221efbc0f583c80258189ab93e4f0d6b6a61ad61335de1fdf3e2331ad423dbeb
MD5 b7280163e09673d256a3f203413973df
BLAKE2b-256 d8e92ea469751ed4df75a911c4bf4dcad1ff5b71ccc9aa1a4aa108583a2a66b5

See more details on using hashes here.

File details

Details for the file claude_agent_sdk-0.2.148-py3-none-manylinux_2_17_aarch64.whl.

File metadata

File hashes

Hashes for claude_agent_sdk-0.2.148-py3-none-manylinux_2_17_aarch64.whl
Algorithm Hash digest
SHA256 3fbbea7cc30099d14f894e12fef68b0ee42fa9982567df94e59499b6aff55959
MD5 9fc470ea3bf5a5b360f40bfeaabedb98
BLAKE2b-256 34d453de4727576fa6b5a261414d420895f4eb5f0369e232d403f1096205709e

See more details on using hashes here.

File details

Details for the file claude_agent_sdk-0.2.148-py3-none-macosx_11_0_x86_64.whl.

File metadata

File hashes

Hashes for claude_agent_sdk-0.2.148-py3-none-macosx_11_0_x86_64.whl
Algorithm Hash digest
SHA256 c8cca1f4e4eb4e0c9f44ccfca6f651cf35e8f3914e483632ae8867c5619818a0
MD5 55479d1a4e4245cc51616ade8d5d2ef9
BLAKE2b-256 afcfe1b11dd9fac877c8b44f022cf895f54f23854b916e09b32360e71209904c

See more details on using hashes here.

File details

Details for the file claude_agent_sdk-0.2.148-py3-none-macosx_11_0_arm64.whl.

File metadata

File hashes

Hashes for claude_agent_sdk-0.2.148-py3-none-macosx_11_0_arm64.whl
Algorithm Hash digest
SHA256 026e2b54bb3f3909712ff462b8469e2b1fc5471cfde796abc23f32d30b2c8ba1
MD5 d163b5c2958f95f9ebef81c8413cef2a
BLAKE2b-256 b8f5f4affb9f882bcf61422ea48491effb6ee2de28bfbdafd0ec90d1b959dad2

See more details on using hashes here.

Release history Release notifications | RSS feed

This release

0.2.148 This release

6 files

0.2.147

6 files

0.2.146

6 files

0.2.145

6 files

0.2.144

6 files

0.2.143

6 files

0.2.142

6 files

0.2.141

6 files

0.2.140

6 files

0.2.139

6 files

0.2.138

6 files

0.2.137

6 files

0.2.136

6 files

0.2.135

6 files

0.2.134

6 files

0.2.133

6 files

0.2.132

6 files

0.2.131

6 files

0.2.130

6 files

0.2.129

6 files

0.2.128

6 files

0.2.127

6 files

0.2.126

6 files

0.2.125

6 files

0.2.124

6 files

0.2.123

6 files

0.2.122

6 files

0.2.121

6 files

0.2.120

6 files

0.2.119

6 files

0.2.118

6 files

0.2.117

6 files

0.2.116

6 files

0.2.115

6 files

0.2.114

6 files

0.2.113

6 files

0.2.112

6 files

0.2.111

6 files

0.2.110

6 files

0.2.109

6 files

0.2.108

6 files

0.2.107

6 files

0.2.106

6 files

0.2.105

6 files

0.2.104

6 files

0.2.103

6 files

0.2.102

6 files

0.2.101

6 files

0.2.100

6 files

0.2.99

6 files

0.2.98

6 files

0.2.97

6 files

0.2.96

6 files

0.2.95

6 files

0.2.94

6 files

0.2.93

6 files

0.2.92

6 files

0.2.91

6 files

0.2.90

6 files

0.2.89

6 files

0.2.88

6 files

0.2.87

6 files

0.2.86

6 files

0.2.85

6 files

0.2.84

6 files

0.2.83

6 files

0.2.82

6 files

0.1.81

6 files

0.1.80

6 files

0.1.79

6 files

0.1.78

6 files

0.1.77

6 files

0.1.76

6 files

0.1.75

6 files

0.1.74

6 files

0.1.73

6 files

0.1.72

6 files

0.1.71

6 files

0.1.70

6 files

0.1.69

6 files

0.1.68

6 files

0.1.67

6 files

0.1.66

6 files

0.1.65

6 files

0.1.64

6 files

0.1.63

6 files

0.1.62

6 files

0.1.61

6 files

0.1.60

6 files

0.1.59

6 files

0.1.58

6 files

0.1.57

6 files

0.1.56

6 files

0.1.55

6 files

0.1.54

6 files

0.1.53

6 files

0.1.52

6 files

0.1.51

6 files

0.1.50

6 files

0.1.49

6 files

0.1.48

5 files

0.1.47

5 files

0.1.46

5 files

0.1.45

5 files

0.1.44

5 files

0.1.43

5 files

0.1.42

5 files

0.1.41

5 files

0.1.40

5 files

0.1.39

5 files

0.1.38

5 files

0.1.37

5 files

0.1.36

5 files

0.1.35

5 files

0.1.34

5 files

0.1.33

5 files

0.1.32

5 files

0.1.31

5 files

0.1.30

5 files

0.1.29

5 files

0.1.28

5 files

0.1.27

5 files

0.1.26

5 files

0.1.25

5 files

0.1.24

5 files

0.1.23

5 files

0.1.22

5 files

0.1.21

5 files

0.1.20

5 files

0.1.19

5 files

0.1.18

5 files

0.1.17

5 files

0.1.16

5 files

0.1.15

5 files

0.1.14

5 files

0.1.13

5 files

0.1.12

5 files

0.1.11

5 files

0.1.10

4 files

0.1.9

4 files

0.1.8

4 files

0.1.7

2 files

0.1.6

2 files

0.1.5

2 files

0.1.4

2 files

0.1.3

2 files

0.1.2

2 files

0.1.1

2 files

0.1.0

2 files

0.0.23

2 files

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page