Bundle platform-specific Claude Code CLI binaries directly in the Python
package, eliminating the need for separate CLI installation.
## Changes
### Build System
- Created `scripts/download_cli.py` to fetch CLI during build
- Created `scripts/build_wheel.py` for building platform-specific wheels
- Created `scripts/update_cli_version.py` to track bundled CLI version
- Updated `pyproject.toml` to properly bundle CLI without duplicate file
warnings
- Made twine check non-blocking (License-File warnings are false
positives)
### Runtime
- Modified `subprocess_cli.py` to check for bundled CLI first
- Added `_cli_version.py` to track which CLI version is bundled
- SDK automatically uses bundled CLI, falling back to system
installation if not found
- Users can still override with `cli_path` option
### Release Workflow
- Updated GitHub workflow to build separate wheels per platform (macOS,
Linux, Windows)
- Workflow now accepts two inputs:
- `version`: Package version to publish (e.g., `0.1.5`)
- `claude_code_version`: CLI version to bundle (e.g., `2.0.0` or
`latest`)
- Workflow builds platform-specific wheels with bundled CLI
- Creates release PR that updates:
- `pyproject.toml` version
- `src/claude_agent_sdk/_version.py`
- `src/claude_agent_sdk/_cli_version.py` with bundled CLI version
- `CHANGELOG.md` with auto-generated release notes
### Documentation
- Updated README to reflect bundled CLI (removed Node.js requirement)
- Added release workflow documentation
- Added local wheel building instructions
## Benefits
- **Zero external dependencies**: No need for Node.js or npm
- **Easier installation**: Single `pip install` gets everything
- **Version control**: Track exactly which CLI version is bundled
- **Flexible releases**: Can release new package versions with updated
CLI without code changes
- **Better user experience**: Works out of the box with no setup
Platform-specific wheels are automatically selected by pip during
installation based on the user's OS and architecture.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-authored-by: Claude <noreply@anthropic.com>
This PR updates the version to 0.1.7 after publishing to PyPI.
## Changes
- Updated version in `pyproject.toml`
- Updated version in `src/claude_agent_sdk/_version.py`
- Updated `CHANGELOG.md` with release notes
## Release Information
- Published to PyPI: https://pypi.org/project/claude-agent-sdk/0.1.7/
- Install with: `pip install claude-agent-sdk==0.1.7`
🤖 Generated by GitHub Actions
---------
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
Co-authored-by: claude[bot] <41898282+claude[bot]@users.noreply.github.com>
Co-authored-by: inigo <bogini@users.noreply.github.com>
Add support for automatic model fallback when primary model is
overloaded. The Python SDK passes the fallback_model parameter to the
Claude CLI, which handles the validation and fallback logic.
Changes:
- Add fallback_model parameter to ClaudeAgentOptions
- Pass --fallback-model to CLI subprocess
- Add test for fallback model command building
The validation that fallback_model != model happens at the CLI layer,
keeping the SDK implementation simple and focused on parameter passing.
---------
Co-authored-by: Claude <noreply@anthropic.com>
Add ~/.claude/local/claude to the list of locations checked when finding
the Claude CLI binary.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-authored-by: Claude <noreply@anthropic.com>
This PR updates the version to 0.1.6 after publishing to PyPI.
## Changes
- Updated version in `pyproject.toml`
- Updated version in `src/claude_agent_sdk/_version.py`
- Updated `CHANGELOG.md` with release notes
## Release Information
- Published to PyPI: https://pypi.org/project/claude-agent-sdk/0.1.6/
- Install with: `pip install claude-agent-sdk==0.1.6`
🤖 Generated by GitHub Actions
---------
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
Co-authored-by: claude[bot] <41898282+claude[bot]@users.noreply.github.com>
Co-authored-by: Ashwin Bhat <ashwin@anthropic.com>
Reduce CI job count by only running examples on Python 3.13 instead of
all Python versions (3.10-3.13). This reduces the combinatorial
explosion while still ensuring examples work on the latest Python
version.
Co-authored-by: Claude <noreply@anthropic.com>
Add support for limiting API costs using the max_budget_usd option,
mirroring the TypeScript SDK functionality. When the budget is exceeded,
query execution stops and returns a result with subtype
'error_max_budget_usd'.
- Add max_budget_usd field to ClaudeAgentOptions
- Pass --max-budget-usd flag to Claude Code CLI
- Add test coverage for budget limit behavior
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-authored-by: Claude <noreply@anthropic.com>
Add support for controlling the maximum number of tokens allocated to
extended thinking blocks via the max_thinking_tokens parameter.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-authored-by: Claude <noreply@anthropic.com>
Move changelog generation prompt from publish.yml into a reusable slash
command at .claude/commands/generate-changelog.md. Update workflow to
call the command with version parameters.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-authored-by: Claude <noreply@anthropic.com>
This PR updates the version to 0.1.5 after publishing to PyPI.
## Changes
- Updated version in `pyproject.toml`
- Updated version in `src/claude_agent_sdk/_version.py`
- Updated `CHANGELOG.md` with release notes
## Release Information
- Published to PyPI: https://pypi.org/project/claude-agent-sdk/0.1.5/
- Install with: `pip install claude-agent-sdk==0.1.5`
🤖 Generated by GitHub Actions
---------
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
Co-authored-by: claude[bot] <41898282+claude[bot]@users.noreply.github.com>
Co-authored-by: Claude <noreply@anthropic.com>
Add SdkPluginConfig type and plugins field to ClaudeAgentOptions.
Plugins can be loaded using the local type with a path to the plugin
directory.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-authored-by: Claude <noreply@anthropic.com>
This PR updates the version to 0.1.4 after publishing to PyPI.
## Changes
- Updated version in `pyproject.toml`
- Updated version in `src/claude_agent_sdk/_version.py`
- Updated `CHANGELOG.md` with release notes
## Release Information
- Published to PyPI: https://pypi.org/project/claude-agent-sdk/0.1.4/
- Install with: `pip install claude-agent-sdk==0.1.4`
🤖 Generated by GitHub Actions
---------
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
Co-authored-by: claude[bot] <41898282+claude[bot]@users.noreply.github.com>
Co-authored-by: Ashwin Bhat <ashwin@anthropic.com>
The changelog generation step was failing to get the previous release
tag because the checkout action was doing a shallow clone. Adding
fetch-depth: 0 ensures all tags are available for git describe.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-authored-by: Claude <noreply@anthropic.com>
## Summary
Fixes#238 - Resolves "command line too long" error on Windows when
using multiple subagents with long prompts.
## Problem
On Windows, the command line length is limited to 8191 characters
(cmd.exe). When using multiple subagents with long prompts, the
`--agents` JSON argument can easily exceed this limit, causing the
error:
```
命令行太长。 (command line too long)
Fatal error in message reader: Command failed with exit code 1
```
## Solution
This PR implements automatic detection and handling of command line
length limits:
1. **Platform-specific limits**:
- Windows: 8000 characters (safe margin below 8191)
- Other platforms: 100,000 characters
2. **Automatic fallback**: When the command line would exceed the limit:
- Write agents JSON to a temporary file
- Use Claude CLI's `@filepath` syntax to reference the file
- Clean up temp files when transport is closed
3. **Zero breaking changes**: The fix is transparent to users - it
automatically activates only when needed
## Changes
- Add `platform` and `tempfile` imports
- Add `_CMD_LENGTH_LIMIT` constant with platform-specific values
- Track temporary files in `self._temp_files` list
- Modify `_build_command()` to detect long command lines and use temp
files
- Clean up temp files in `close()` method
## Testing
- ✅ All existing tests pass (122 tests)
- ✅ Linting and type checking pass
- ✅ Minimal changes - only 47 lines added/modified
- ✅ Solution transparently handles the Windows command line limit
## Test plan
- [x] Test on Windows with multiple subagents and long prompts
- [x] Verify temp files are created and cleaned up properly
- [x] Verify normal operation (short command lines) is unaffected
- [x] Test cross-platform compatibility (limit only applies on Windows)
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-authored-by: Claude <noreply@anthropic.com>
Adds environment variable to allow skipping the Claude Code version
check. Users can set CLAUDE_AGENT_SDK_SKIP_VERSION_CHECK to disable the
check.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-authored-by: Claude <noreply@anthropic.com>
With this change claude_code_sdk tools can now also return base64
images, in addition to text, following the MCP standard.
For example:
```
import base64
@tool("make_response", "prompt", {"foo": str})
def make_response(args):
fake_bytes = b"someimagebytes"
encoded_image = base64.b64encode(fake_bytes).decode("utf-8")
return {
"content": [
{"type": "text", "text": "Hello world!"},
{
"type": "image",
"mimeType": "image/jpeg",
"data": encoded_image,
},
]
}
```
The image will now be sent to Claude Code when calling the function,
allowing it to react to it.
---------
Co-authored-by: Ashwin Bhat <ashwin@anthropic.com>
Adds a committable pre-push hook that runs ruff checks before pushing,
matching the CI lint workflow. Developers run ./scripts/initial-setup.sh
to install the hook locally.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-authored-by: Claude <noreply@anthropic.com>
## Summary
Adds support for passing custom Claude Code CLI paths through
`ClaudeAgentOptions`, allowing organizations with non-standard
installation locations to specify the CLI path explicitly.
## Motivation
As noted in #214, organizations may install Claude Code CLI (or wrapped
versions) at custom locations and prefer to provide those paths instead
of relying on the SDK's default search logic. The transport layer
already supported `cli_path`, but it was never exposed through the
public API.
## Changes
1. **types.py**: Added `cli_path: str | Path | None = None` parameter to
`ClaudeAgentOptions` dataclass
2. **_internal/client.py**: Pass `cli_path` from
`configured_options.cli_path` to `SubprocessCLITransport`
3. **client.py**: Pass `cli_path` from `options.cli_path` to
`SubprocessCLITransport`
## Implementation Details
The `SubprocessCLITransport` constructor already accepted a `cli_path`
parameter (line 40 of subprocess_cli.py), but it was never passed from
the client layers. This PR completes the wiring by:
- Adding the option to the public `ClaudeAgentOptions` interface
- Extracting and passing it through both client implementations
(`InternalClient.process_query` and `ClaudeSDKClient.connect`)
## Usage Example
```python
from claude_agent_sdk import query, ClaudeAgentOptions
# Specify custom CLI path
options = ClaudeAgentOptions(
cli_path="/custom/path/to/claude"
)
result = await query("Hello!", options=options)
```
## Testing
- No new tests added as this is a straightforward parameter pass-through
- Existing tests should continue to work (default behavior unchanged)
- CI will validate the changes don't break existing functionality
Fixes#214
---------
Co-authored-by: Ashwin Bhat <ashwin@anthropic.com>
This PR updates the version to 0.1.3 after publishing to PyPI.
## Changes
- Updated version in `pyproject.toml`
- Updated version in `src/claude_agent_sdk/_version.py`
- Updated `CHANGELOG.md` with release notes
## Release Information
- Published to PyPI: https://pypi.org/project/claude-agent-sdk/0.1.3/
- Install with: `pip install claude-agent-sdk==0.1.3`
🤖 Generated by GitHub Actions
---------
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
Co-authored-by: claude[bot] <41898282+claude[bot]@users.noreply.github.com>
Co-authored-by: Ashwin Bhat <ashwin@anthropic.com>
Add typed hook input structures (PreToolUseHookInput,
PostToolUseHookInput, etc.) to provide better IDE autocomplete and type
safety for hook callbacks. Also convert HookContext from dataclass to
TypedDict to match runtime behavior.
Changes:
- Add BaseHookInput, PreToolUseHookInput, PostToolUseHookInput,
UserPromptSubmitHookInput, StopHookInput, SubagentStopHookInput, and
PreCompactHookInput TypedDict classes
- Update HookCallback signature to use HookInput union type
- Convert HookContext from dataclass to TypedDict (fixes type mismatch)
- Export all new hook input types from __init__.py
- Update all examples and tests to use typed hook inputs
Benefits:
- Zero breaking changes (TypedDict is dict-compatible at runtime)
- Full type safety and IDE autocomplete for hook callbacks
- Matches TypeScript SDK structure exactly
- Self-documenting hook input fields
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-authored-by: Claude <noreply@anthropic.com>
Fixes critical bug where hook outputs using `async_` and `continue_`
(Python-safe names avoiding keyword conflicts) were not being converted
to `async` and `continue` as expected by the CLI. This caused hook
control fields like `{"decision": "block"}` or `{"continue_": False}` to
be silently ignored.
Changes:
- Add _convert_hook_output_for_cli() to handle field name conversion
- Apply conversion in hook callback handling
- Update type documentation to clarify field name usage
- Add comprehensive test coverage for field name conversion
- Update existing tests to verify conversion occurs correctly
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-authored-by: Claude <noreply@anthropic.com>
This PR updates the version to 0.1.2 after publishing to PyPI.
## Changes
- Updated version in `pyproject.toml`
- Updated version in `src/claude_agent_sdk/_version.py`
- Updated `CHANGELOG.md` with release notes
## Release Information
- Published to PyPI: https://pypi.org/project/claude-agent-sdk/0.1.2/
- Install with: `pip install claude-agent-sdk==0.1.2`
🤖 Generated by GitHub Actions
---------
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
Co-authored-by: claude[bot] <41898282+claude[bot]@users.noreply.github.com>
Co-authored-by: Claude <noreply@anthropic.com>
Closes the gap between Python and TypeScript SDK hook output types by
adding:
- `reason` field for explaining decisions
- `continue_` field for controlling execution flow
- `suppressOutput` field for hiding stdout
- `stopReason` field for stop explanations
- `decision` now supports both "approve" and "block" (not just "block")
- `AsyncHookJSONOutput` type for deferred hook execution
- Proper typing for `hookSpecificOutput` with discriminated unions
Also adds comprehensive examples and tests:
- New examples in hooks.py demonstrating all new fields
- Unit tests in test_tool_callbacks.py for new output types
- E2E tests in e2e-tests/test_hooks.py with real API calls
- CI integration in .github/workflows/test.yml
🤖 Generated with [Claude Code](https://claude.com/claude-code)
---------
Co-authored-by: Claude <noreply@anthropic.com>
Temporarily removes windows-latest from the test-e2e job matrix to
disable Windows end-to-end testing. Unit tests continue to run on
Windows.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-authored-by: Claude <noreply@anthropic.com>
https://github.com/anthropics/claude-agent-sdk-python/issues/227
Fixes zod issues returned by cli subprocess transport - example now runs
correctly to completion:
```
============================================================
Tool Permission Callback Example
============================================================
This example demonstrates how to:
1. Allow/deny tools based on type
2. Modify tool inputs for safety
3. Log tool usage
4. Prompt for unknown tools
============================================================
📝 Sending query to Claude...
📨 Receiving response...
💬 Claude: I'll help you with these tasks. Let me execute them in sequence.
💬 Claude: Now I'll create a simple Python hello world script:
🔧 Tool Permission Request: Write
Input: {
"file_path": "/Users/vimota/code/claude-agent-sdk-python/hello.py",
"content": "#!/usr/bin/env python3\n\nprint(\"Hello, World!\")\n"
}
⚠️ Redirecting write from /Users/vimota/code/claude-agent-sdk-python/hello.py to ./safe_output/hello.py
💬 Claude: Now let's run the script:
🔧 Tool Permission Request: Bash
Input: {
"command": "python hello.py",
"description": "Run hello.py script"
}
✅ Allowing bash command: python hello.py
💬 Claude: Let me check where the file was created:
💬 Claude: I see the file was created in the `safe_output` directory. Let me run it from there:
🔧 Tool Permission Request: Bash
Input: {
"command": "python ./safe_output/hello.py",
"description": "Run hello.py from safe_output"
}
✅ Allowing bash command: python ./safe_output/hello.py
💬 Claude: Perfect! All tasks completed successfully:
1. **Listed files** - The directory contains a Python SDK project with source code in `src/`, tests, examples, and configuration files.
2. **Created hello.py** - A simple Python script was created at `./safe_output/hello.py` with a basic "Hello, World!" print statement.
3. **Ran the script** - The script executed successfully and printed "Hello, World!" to the console.
Note: The file was created in the `safe_output/` subdirectory rather than the root directory.
✅ Task completed!
Duration: 31158ms
Cost: $0.0736
Messages processed: 18
============================================================
Tool Usage Summary
============================================================
1. Tool: Write
Input: {
"file_path": "/Users/vimota/code/claude-agent-sdk-python/hello.py",
"content": "#!/usr/bin/env python3\n\nprint(\"Hello, World!\")\n"
}
Suggestions: [{'type': 'setMode', 'mode': 'acceptEdits', 'destination': 'session'}]
2. Tool: Bash
Input: {
"command": "python hello.py",
"description": "Run hello.py script"
}
Suggestions: [{'type': 'addRules', 'rules': [{'toolName': 'Bash', 'ruleContent': 'python:*'}], 'behavior': 'allow', 'destination': 'localSettings'}]
3. Tool: Bash
Input: {
"command": "python ./safe_output/hello.py",
"description": "Run hello.py from safe_output"
}
Suggestions: [{'type': 'addRules', 'rules': [{'toolName': 'Bash', 'ruleContent': 'python:*'}], 'behavior': 'allow', 'destination': 'localSettings'}]
```
Replace GitHub API-based commits with local git workflow and integrate
claude-code-action to automatically generate changelog entries. The
workflow now:
- Creates release branch locally with version commits
- Uses Claude to review changes and update CHANGELOG.md
- Pushes complete branch with all commits together
🤖 Generated with [Claude Code](https://claude.com/claude-code)
---------
Co-authored-by: Claude <noreply@anthropic.com>
Document the features and improvements released in version 0.1.1,
including minimum Claude Code version check, updated PermissionResult
types, and simplified model references.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-authored-by: Claude <noreply@anthropic.com>
Add cross-platform testing for Windows alongside Linux across all test
jobs (unit tests, e2e tests, and examples). Uses native Windows
installation via PowerShell script and platform-specific timeout
handling for example scripts.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
---------
Co-authored-by: Claude <noreply@anthropic.com>
- Make test_cli_path_accepts_pathlib_path platform-aware by comparing
with str(path)
- Fix test_build_command_with_add_dirs to check directories individually
- Fix test_build_command_with_mcp_servers_as_file_path to handle Path
conversion
- Fix test_query_with_async_iterable to properly execute Python scripts
on Windows
All tests now pass on both Windows and Unix-like systems (110/110 tests
passing).
Fixes the issue #217 where pathlib.Path automatically converts path
separators based on the operating system, causing test assertions to
fail on Windows.
This PR updates the version to 0.1.1 after publishing to PyPI.
## Changes
- Updated version in `pyproject.toml`
- Updated version in `src/claude_agent_sdk/_version.py`
## Release Information
- Published to PyPI: https://pypi.org/project/claude-agent-sdk/0.1.1/
- Install with: `pip install claude-agent-sdk==0.1.1`
🤖 Generated by GitHub Actions
---------
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
- Add version check in subprocess transport that runs `claude -v` on
connect
- Display warning to stderr if version is below 2.0.0
- Update README prerequisites to specify Claude Code 2.0.0+
- Version check is non-blocking (warns but continues execution)
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-authored-by: Claude <noreply@anthropic.com>
Replace all dated model references (claude-sonnet-4-20250514,
claude-3-5-sonnet-20241022) with the simplified claude-sonnet-4-5
identifier across examples, tests, and documentation.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-authored-by: Claude <noreply@anthropic.com>
`query` supports custom transport, but ClaudeSDKClient does not. So I
add this feature. eg.
```python
from claude_code_sdk import ClaudeSDKClient, Transport
class MyCustomTransport(Transport):
# Implement custom transport logic
pass
transport = MyCustomTransport()
async with ClaudeSDKClient( transport=transport) as client:
await client.query("Greet Alice")
# Extract and print response
async for msg in client.receive_response():
print(msg)
```
This PR updates the version to 0.1.0 after publishing to PyPI.
## Changes
- Updated version in `pyproject.toml`
- Updated version in `src/claude_agent_sdk/_version.py`
## Release Information
- Published to PyPI: https://pypi.org/project/claude-agent-sdk/0.1.0/
- Install with: `pip install claude-agent-sdk==0.1.0`
🤖 Generated by GitHub Actions
---------
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
## Summary
- Adds `set_permission_mode()` method to dynamically change permission
modes during streaming sessions
- Adds `set_model()` method to switch AI models mid-conversation
- Implements control protocol support matching the TypeScript SDK's
capabilities
## Motivation
The TypeScript SDK supports dynamic control through
`setPermissionMode()` and `setModel()` methods on the Query interface.
This PR brings the same functionality to the Python SDK, allowing users
to:
1. Start with restrictive permissions for code review, then switch to
`acceptEdits` for implementation
2. Use different models for different parts of a task (e.g., Sonnet for
complex analysis, Haiku for simple tasks)
3. Adjust permissions based on workflow needs without restarting
sessions
## Changes
- **ClaudeSDKClient**: Added `set_permission_mode(mode)` and
`set_model(model)` methods
- **Internal Query class**: Added `set_model(model)` method to send
control requests
- **E2E tests**: Added comprehensive tests verifying the functionality
works with real API calls
## Test Plan
- [x] All existing unit tests pass (102 tests)
- [x] New E2E tests added and passing:
- `test_set_permission_mode`: Verifies permission mode changes take
effect
- `test_set_model`: Confirms model switching works mid-conversation
- `test_interrupt`: Validates interrupt capability
- [x] Type checking passes (`mypy`)
- [x] Linting passes (`ruff`)
## Usage Example
```python
async with ClaudeSDKClient() as client:
# Start with default permissions for review
await client.query("Analyze this code for issues")
# Switch to auto-accept edits for implementation
await client.set_permission_mode('acceptEdits')
await client.query("Now fix the issues we found")
# Use a different model for simpler tasks
await client.set_model('claude-3-5-haiku-20241022')
await client.query("Add a simple docstring")
```
🤖 Generated with [Claude Code](https://claude.ai/code)
---------
Co-authored-by: Claude <noreply@anthropic.com>
Co-authored-by: Ashwin Bhat <ashwin@anthropic.com>
Replace separate system_prompt and append_system_prompt fields with a
single system_prompt field that accepts:
- string: custom system prompt
- {"preset": "claude_code"}: use default Claude Code prompt
- {"preset": "claude_code", "append": "..."}: default prompt with
additions
- None/undefined: vanilla Claude with no system prompt
This matches the TypeScript SDK API design and provides more flexible
system prompt configuration.
🤖 Generated with [Claude Code](https://claude.ai/code)
---------
Co-authored-by: Claude <noreply@anthropic.com>
## Summary
- Add support for custom agent definitions via `agents` option
- Add support for controlling setting sources via `setting_sources`
option
- Add `/commit` slash command to project
- Add examples demonstrating both features
- Add e2e tests for verification
## Changes
### Core Implementation
- Add `AgentDefinition` and `SettingSource` types to `types.py`
- Add `agents` and `setting_sources` fields to `ClaudeCodeOptions`
- Update subprocess CLI transport to pass `--agents` and
`--setting-sources` flags
- **Default behavior**: When `setting_sources` is not provided, pass
empty string (no settings loaded)
- Handle empty `setting_sources` array correctly (pass empty string to
CLI)
### Examples
- `examples/agents.py`: Demonstrates custom agent definitions with
different tools and models
- `examples/setting_sources.py`: Shows how setting sources control which
settings are loaded
- Default behavior (no settings)
- User-only settings
- User + project settings
### Tests
- Add e2e tests verifying agents and setting_sources functionality
- Test default behavior (no settings loaded)
- Test filtering by setting source
- Use `output_style` checking to verify settings loaded/not loaded
- Tests use temporary directories for isolated testing
### Project Config
- Add `.claude/commands/commit.md` slash command for git commits
## Test Plan
- [x] E2E tests added for all new functionality
- [ ] CI tests pass
- [ ] Examples run successfully
🤖 Generated with [Claude Code](https://claude.ai/code)
---------
Co-authored-by: Claude <noreply@anthropic.com>
Include the SDK version in the environment when spawning the Claude CLI
process.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-authored-by: Claude <noreply@anthropic.com>
Add support for fork_session parameter which allows resumed sessions to
fork to a new session ID rather than continuing the previous session.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-authored-by: Claude <noreply@anthropic.com>
