Skip to content

CLAUDE.md

Latest commit

 

History

History
61 lines (43 loc) · 1.62 KB

CLAUDE.md

File metadata and controls

61 lines (43 loc) · 1.62 KB

CLAUDE.md

Guidance for Claude Code when working on this repository.

What This Project Is

pydantic-ai-todo provides a todo/task planning toolset for pydantic-ai agents. It's designed to work standalone with any agent - no specific deps type required.

Key pattern: Storage Injection - tools use closure to capture state, not ctx.deps.

Commands

uv sync --group dev          # Install dependencies
uv run pytest                # Run tests
uv run coverage run -m pytest && uv run coverage report  # Test with coverage
uv run ruff check .          # Lint
uv run ruff format .         # Format
uv run pyright               # Type check

Structure

pydantic_ai_todo/
├── types.py      # Todo, TodoItem models
├── storage.py    # TodoStorage, TodoStorageProtocol
├── toolset.py    # create_todo_toolset(), get_todo_system_prompt()
└── __init__.py   # Public API exports

Core Pattern

def create_todo_toolset(storage=None) -> FunctionToolset[Any]:
    _storage = storage or TodoStorage()

    @toolset.tool
    async def read_todos() -> str:  # No ctx - uses closure
        return format(_storage.todos)

Requirements

  • 100% test coverage - every PR must maintain this
  • Type annotations - pyright strict mode
  • Async tools - all tool functions are async

Testing

# Run specific test
uv run pytest tests/test_toolset.py::TestCreateTodoToolset -v

# Debug mode
uv run pytest -v -s

Integration

This library is used by pydantic-deep which re-exports its API. Changes here affect pydantic-deep users.