📄 CodeRabbit inference engine (CLAUDE.md)

All Python dependency versions must use == for pinned versions in [project.dependencies] and [tool.uv.sources]

Files:

• pyproject.toml

📄 CodeRabbit inference engine (CLAUDE.md)

**/*.py: No from __future__ import annotations — Python 3.14 has PEP 649 native lazy annotations
Use except A, B: syntax (no parentheses) following PEP 758 — ruff enforces this on Python 3.14
All public functions and classes must have type hints — mypy strict mode is enforced
Docstrings must follow Google style format and are required on all public classes and functions — enforced by ruff D rules
Line length must be 88 characters — enforced by ruff
Functions must be fewer than 50 lines and files must be fewer than 800 lines
Handle errors explicitly — never silently swallow exceptions
Validate at system boundaries (user input, external APIs, config files)
Use Pydantic v2 with BaseModel, model_validator, computed_field, and ConfigDict
Use @computed_field for derived values instead of storing and validating redundant fields (e.g., TokenUsage.total_tokens)
Use NotBlankStr (from core.types) for all identifier and name fields — including optional (NotBlankStr | None) and tuple (tuple[NotBlankStr, ...]) variants — instead of manual whitespace validators
Prefer asyncio.TaskGroup for fan-out/fan-in parallel operations in new code (e.g., multiple tool invocations, parallel agent calls) — use structured concurrency over bare create_task
Create new objects rather than mutating existing ones — use copy.deepcopy() at construction and MappingProxyType wrapping for read-only enforcement on non-Pydantic internal collections
Use frozen Pydantic models for config and identity; use separate mutable-via-copy models (with model_copy(update=...)) for runtime state that evolves — never mix static config fields with mutable runtime fields in one model
All provider calls must go through BaseCompletionProvider which applies retry and rate limiting automatically — never implement retry logic in driver subclasses or calling code
RetryConfig and RateLimiterConfig are set per-provider in ProviderConfig — retryable errors (RateLimitError, `ProviderTime...

Files:

• src/synthorg/__init__.py

📄 CodeRabbit inference engine (CLAUDE.md)

src/synthorg/**/*.py: Every module with business logic must import logger via from synthorg.observability import get_logger and create logger = get_logger(__name__)
Never use import logging, logging.getLogger(), or print() in application code — use the logger from synthorg.observability
Logger variable must always be named logger (not _logger or log)
Event names must always use constants from domain-specific modules under synthorg.observability.events (e.g., API_REQUEST_STARTED from events.api, TOOL_INVOKE_START from events.tool)
Use structured logging with keyword arguments: logger.info(EVENT, key=value) — never use format strings like logger.info('msg %s', val)
All error paths must log at WARNING or ERROR level with context before raising an exception
All state transitions must log at INFO level
DEBUG logging should be used for object creation, internal flow, and entry/exit of key functions
Pure data models, enums, and re-exports do NOT need logging

Files:

• src/synthorg/__init__.py