Skip to content

Compress tool outputs, logs, files, and RAG chunks before they reach the LLM. 20% fewer tokens for coding agents, 60-95% fewer tokens for JSON, same answers. Library, proxy, MCP server.

fix: type cache lazy exports

JerrettDavis245241d2b486
diff --git a/headroom/__init__.py b/headroom/__init__.py
index c7061b3f1..8e65f3750 100644
--- a/headroom/__init__.py
+++ b/headroom/__init__.py
@@ -71,6 +71,7 @@ For more examples, see https://github.com/headroom-sdk/headroom/tree/main/exampl
from __future__ import annotations
from importlib import import_module
+from typing import Any
from ._version import __version__ # noqa: F401
from .compress import CompressResult, compress
@@ -275,7 +276,7 @@ _OPTIONAL_EXPORTS = {
}
-def __getattr__(name: str):
+def __getattr__(name: str) -> Any:
module_attr = _LAZY_EXPORTS.get(name)
if module_attr is not None:
module_name, attr_name = module_attr
diff --git a/headroom/cache/__init__.py b/headroom/cache/__init__.py
index f9899deff..88315247c 100644
--- a/headroom/cache/__init__.py
+++ b/headroom/cache/__init__.py
@@ -3,6 +3,38 @@
from __future__ import annotations
from importlib import import_module
+from typing import TYPE_CHECKING
+
+if TYPE_CHECKING:
+ from headroom.cache.anthropic import AnthropicCacheOptimizer # noqa: F401
+ from headroom.cache.base import ( # noqa: F401
+ BaseCacheOptimizer,
+ CacheBreakpoint,
+ CacheConfig,
+ CacheMetrics,
+ CacheOptimizer,
+ CacheResult,
+ CacheStrategy,
+ OptimizationContext,
+ )
+ from headroom.cache.compression_cache import CompressionCache # noqa: F401
+ from headroom.cache.dynamic_detector import ( # noqa: F401
+ DetectorConfig,
+ DynamicCategory,
+ DynamicContentDetector,
+ DynamicSpan,
+ detect_dynamic_content,
+ )
+ from headroom.cache.google import GoogleCacheOptimizer # noqa: F401
+ from headroom.cache.openai import OpenAICacheOptimizer # noqa: F401
+ from headroom.cache.prefix_tracker import ( # noqa: F401
+ FreezeStats,
+ PrefixCacheTracker,
+ PrefixFreezeConfig,
+ SessionTrackerStore,
+ )
+ from headroom.cache.registry import CacheOptimizerRegistry # noqa: F401
+ from headroom.cache.semantic import SemanticCache, SemanticCacheLayer # noqa: F401
__all__ = [
# Base types