agentbyte 0.52.0__tar.gz → 0.53.0__tar.gz

{agentbyte-0.52.0 → agentbyte-0.53.0}/CHANGELOG.md

@@ -6,6 +6,46 @@ The format follows Keep a Changelog principles and semantic versioning.
 
 ## [Unreleased]
 
+## [0.53.0] - 2026-10-10
+
+### Added
+
+- `ClientMiddleware`: a marker subclass of `BaseMiddleware` for middleware safe to attach
+  directly to a client — no `AgentContext` or conversational state required. `OTelMiddleware`,
+  `LoggingMiddleware`, `MetricsMiddleware`, and `RateLimitMiddleware` now qualify. Every
+  embedding client (`OpenAIEmbeddingClient`, `AzureOpenAIEmbeddingClient`,
+  `CohereBedrockEmbeddingClient`) and every chat client (`OpenAIChatCompletionClient`,
+  `AzureOpenAIChatCompletionClient`, `AnthropicChatCompletionClient`,
+  `AnthropicBedrockChatCompletionClient`) accepts a `middlewares=[...]` constructor argument,
+  getting OTel tracing/logging on direct calls with no `Agent` wrapper required. Passing a
+  plain `BaseMiddleware` that needs agent state (e.g. `ApprovalMiddleware`,
+  `ContextCompactionMiddleware`) raises `TypeError` immediately. Chat clients wrap only the
+  non-streaming `create()`; `create_stream()` remains unwrapped (documented gap for a future
+  spec — see `docs/specs/0075-client-level-middleware/`).
+- `agentbyte.llm.EmbeddingBatcher`: batching/concurrency orchestration over any embedding
+  client (sequential or semaphore-bounded concurrent batches, `run()`/`run_stream()` with
+  `EmbeddingBatchProgressEvent` progress).
+- `BaseEmbeddingClient.middleware_types` / `BaseChatCompletionClient.middleware_types`: a
+  public, read-only property exposing the concrete middleware types attached to a client.
+  `Agent` uses it to warn (`warnings.warn`, not an error) when it and its `model_client` both
+  configure the same middleware class — that call would otherwise run through two nested
+  `MiddlewareChain`s, double-emitting spans/logs/metrics for one real API call. See
+  `docs/design-guide/01-agent_topics/topic_middleware.md`'s "Client-Level Middleware" caveat.
+
+### Changed (breaking)
+
+- Removed `agentbyte.agents.EmbeddingAgent`. It bundled two unrelated concerns — batching and
+  middleware/OTel access — under a misleading "agent" name (no reasoning, no tool calls, no
+  conversational state). Middleware now lives on the embedding client itself (see
+  `ClientMiddleware` above); batching moved to `agentbyte.llm.EmbeddingBatcher`, which drops the
+  `middlewares` and `context: AgentContext` parameters entirely since the client it wraps owns
+  that chain. No compatibility shim. Migration guide:
+  `docs/design-guide/01-agent_topics/topic_embedding_batcher.md`.
+- `EmbeddingBatchProgressEvent` moved from `agentbyte.agents.types` to `agentbyte.llm.types` (now
+  a standalone `pydantic.BaseModel`, not an `agents.types.BaseEvent` subclass) — it was never an
+  agent event, and keeping it in `agents.types` would have made `agentbyte.llm` depend on
+  `agentbyte.agents`, inverting the package layering.
+
 ## [0.52.0] - 2026-10-10
 
 ### Added

{agentbyte-0.52.0 → agentbyte-0.53.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.5
 Name: agentbyte
-Version: 0.52.0
+Version: 0.53.0
 Summary: A toolkit for designing multiagent systems
 Author-email: MrDataPsycho <mr.data.psycho@gmail.com>
 License-Expression: LicenseRef-Proprietary
@@ -90,7 +90,7 @@ Description-Content-Type: text/markdown
 
 Agentbyte is an observability-first agentic AI framework for building and studying multiagent systems with a learning-first, implementation-oriented workflow.
 
-Current release: **0.51.0**
+Current release: **0.53.0**
 
 ## Building an Agent
 
@@ -258,6 +258,8 @@ Other orchestrators follow the same `run()`/`run_stream()` shape: `AIOrchestrato
 
 ## Middleware
 
+Chat and embedding clients also accept `middlewares=[...]` directly for client-safe middleware.
+
 Built in: `LoggingMiddleware`, `PIIRedactionMiddleware`, `GuardrailMiddleware`, `MetricsMiddleware`, `RateLimitMiddleware`, `ApprovalMiddleware`, `ContextCompactionMiddleware`, `RetryMiddleware`, `OTelMiddleware`. Attach any combination to the same support agent via `middlewares=[...]`:
 
 ```python

{agentbyte-0.52.0 → agentbyte-0.53.0}/README.md

@@ -2,7 +2,7 @@
 
 Agentbyte is an observability-first agentic AI framework for building and studying multiagent systems with a learning-first, implementation-oriented workflow.
 
-Current release: **0.51.0**
+Current release: **0.53.0**
 
 ## Building an Agent
 
@@ -170,6 +170,8 @@ Other orchestrators follow the same `run()`/`run_stream()` shape: `AIOrchestrato
 
 ## Middleware
 
+Chat and embedding clients also accept `middlewares=[...]` directly for client-safe middleware.
+
 Built in: `LoggingMiddleware`, `PIIRedactionMiddleware`, `GuardrailMiddleware`, `MetricsMiddleware`, `RateLimitMiddleware`, `ApprovalMiddleware`, `ContextCompactionMiddleware`, `RetryMiddleware`, `OTelMiddleware`. Attach any combination to the same support agent via `middlewares=[...]`:
 
 ```python

agentbyte-0.53.0/src/agentbyte/__about__.py

@@ -0,0 +1,2 @@
+__version__ = "0.53.0"
+VERSION = __version__

{agentbyte-0.52.0 → agentbyte-0.53.0}/src/agentbyte/agents/__init__.py

@@ -11,13 +11,11 @@ from .base import (
     AgentToolError,
     BaseAgent,
 )
-from .embedding_agent import EmbeddingAgent
 from .types import (
     AgentErrorDetail,
     AgentEvent,
     AgentResponse,
     BaseEvent,
-    EmbeddingBatchProgressEvent,
     ErrorEvent,
     FatalErrorEvent,
     MemoryRetrievalEvent,
@@ -37,7 +35,6 @@ __all__ = [
     "BaseAgent",
     "Agent",
     "AgentAsTool",
-    "EmbeddingAgent",
     "AgentError",
     "AgentConfigurationError",
     "AgentExecutionError",
@@ -57,7 +54,6 @@ __all__ = [
     "MemoryRetrievalEvent",
     "ErrorEvent",
     "FatalErrorEvent",
-    "EmbeddingBatchProgressEvent",
     "AgentResponse",
     "AgentErrorDetail",
 ]

{agentbyte-0.52.0 → agentbyte-0.53.0}/src/agentbyte/agents/base.py

@@ -3,6 +3,7 @@ Base agent abstraction and shared helper behavior.
 """
 
 import json
+import warnings
 from abc import ABC, abstractmethod
 from collections.abc import AsyncGenerator, Callable
 from typing import Any, Dict, List, Optional, Type, Union
@@ -98,6 +99,37 @@ class BaseAgent(ABC):
         self.extra_config = kwargs
 
         self._validate_configuration()
+        self._warn_on_duplicate_client_middleware()
+
+    def _warn_on_duplicate_client_middleware(self) -> None:
+        """Warn when this agent and its model_client both configure the same
+        middleware type.
+
+        A client's own ``middlewares`` already wraps its provider call in a
+        ``MiddlewareChain``; this agent wraps that same call in a second chain.
+        Attaching the same middleware class at both layers runs it twice per
+        logical call — e.g. two nested OTel spans and the token/duration
+        histograms recorded twice for one real API call. This is sometimes
+        intentional (e.g. two differently configured rate limiters), so it is
+        a warning, not an error — see ``topic_middleware.md``'s "Client-Level
+        Middleware" section for the full caveat.
+        """
+        client_types = getattr(self.model_client, "middleware_types", None)
+        if not client_types:
+            return
+        agent_types = {type(m) for m in self.middleware_chain.middlewares}
+        overlap = agent_types & client_types
+        if not overlap:
+            return
+        names = ", ".join(sorted(t.__name__ for t in overlap))
+        warnings.warn(
+            f"Agent '{self.name}' and its model_client "
+            f"({type(self.model_client).__name__}) both configure: {names}. "
+            "Each model call will run through both middleware chains, "
+            "double-emitting spans/logs/metrics for one real API call. "
+            "Configure each middleware at one layer only.",
+            stacklevel=3,
+        )
 
     def _validate_configuration(self) -> None:
         """Validate required configuration fields."""

{agentbyte-0.52.0 → agentbyte-0.53.0}/src/agentbyte/agents/types.py

@@ -164,16 +164,6 @@ class FatalErrorEvent(BaseEvent):
     )
 
 
-class EmbeddingBatchProgressEvent(BaseEvent):
-    """Event emitted after each batch completes during an EmbeddingAgent run."""
-
-    event_type: Literal["embedding_batch_progress"] = "embedding_batch_progress"
-    batch_index: int = Field(..., description="Zero-based index of the completed batch")
-    total_batches: int = Field(..., description="Total number of batches for this run")
-    completed_texts: int = Field(..., description="Cumulative number of texts embedded so far")
-    total_texts: int = Field(..., description="Total number of texts in this run")
-
-
 class AgentErrorDetail(BaseModel):
     """Structured detail for a run that ended with ``finish_reason="error"``.
 
@@ -350,7 +340,6 @@ AgentEvent = Union[
     MemoryRetrievalEvent,
     ErrorEvent,
     FatalErrorEvent,
-    EmbeddingBatchProgressEvent,
 ]
 
 
@@ -370,7 +359,6 @@ __all__ = [
     "ModelStreamChunkEvent",
     "ErrorEvent",
     "FatalErrorEvent",
-    "EmbeddingBatchProgressEvent",
     "AgentResponse",
     "AgentErrorDetail",
 ]

{agentbyte-0.52.0 → agentbyte-0.53.0}/src/agentbyte/llm/__init__.py

@@ -33,6 +33,7 @@ from .base import (
 )
 from .retry_policy import RetryPolicy
 from .embeddings_base import BaseEmbeddingClient, BaseEmbeddingClientConfig
+from .embedding_batcher import EmbeddingBatcher
 # Import message types and usage from root module (picoagents-aligned)
 from agentbyte.messages import (
     BaseMessage,
@@ -49,6 +50,7 @@ from agentbyte.messages import (
 from .types import (
     ChatCompletionChunk,
     ChatCompletionResult,
+    EmbeddingBatchProgressEvent,
     EmbeddingResult,
     ModelClientError,
     ModelPricing,
@@ -60,6 +62,8 @@ __all__ = [
     "BaseChatCompletionClientConfig",
     "BaseEmbeddingClient",
     "BaseEmbeddingClientConfig",
+    "EmbeddingBatcher",
+    "EmbeddingBatchProgressEvent",
     # Provider implementations
     "OpenAIChatCompletionClient",
     "OpenAIChatCompletionClientConfig",

{agentbyte-0.52.0 → agentbyte-0.53.0}/src/agentbyte/llm/anthropic/_conversion.py

@@ -47,14 +47,15 @@ DEFAULT_MAX_TOKENS = 4096
 class AnthropicConversionMixin:
     """Messages API request/response conversion shared by both Claude transports."""
 
-    async def create(
+    async def _execute_chat_operation(
         self,
-        messages: List[Message],
-        tools: Optional[List[Dict[str, Any]]] = None,
-        output_format: Optional[type] = None,
-        **kwargs: Any,
+        payload: Dict[str, Any],
     ) -> ChatCompletionResult:
         """Make a single Anthropic Messages API call."""
+        messages = payload["messages"]
+        tools = payload["tools"]
+        output_format = payload["output_format"]
+        kwargs = payload["kwargs"]
         retry_history: List = []
         try:
             result = await self._retry_with_backoff(

{agentbyte-0.52.0 → agentbyte-0.53.0}/src/agentbyte/llm/anthropic/chat.py

@@ -2,11 +2,12 @@
 Anthropic Claude chat completion client implementation (direct API).
 """
 
-from typing import Any, Dict, Optional
+from typing import Any, Dict, List, Optional
 
 from anthropic import AsyncAnthropic
 
 from agentbyte.component import Component
+from agentbyte.middleware.base import ClientMiddleware
 
 from ..base import BaseChatCompletionClient, BaseChatCompletionClientConfig
 from ..pricing import PricingRegistry
@@ -44,8 +45,9 @@ class AnthropicChatCompletionClient(
         initial_retry_delay: float = 1.0,
         max_retry_delay: float = 60.0,
         retry_policy: Optional[RetryPolicy] = None,
+        middlewares: Optional[List[ClientMiddleware]] = None,
     ):
-        super().__init__(model, client, config)
+        super().__init__(model, client, config, middlewares=middlewares)
         self.model_pricing = model_pricing
         if retry_policy is not None:
             self.retry_policy = retry_policy
@@ -80,6 +82,7 @@ class AnthropicChatCompletionClient(
         initial_retry_delay: float = 1.0,
         max_retry_delay: float = 60.0,
         retry_policy: Optional[RetryPolicy] = None,
+        middlewares: Optional[List[ClientMiddleware]] = None,
     ) -> "AnthropicChatCompletionClient":
         """Create a client from an explicit or env-backed API key."""
         from pydantic import ValidationError
@@ -110,6 +113,7 @@ class AnthropicChatCompletionClient(
             initial_retry_delay=initial_retry_delay,
             max_retry_delay=max_retry_delay,
             retry_policy=retry_policy,
+            middlewares=middlewares,
         )
 
     def _estimate_cost(

{agentbyte-0.52.0 → agentbyte-0.53.0}/src/agentbyte/llm/azure/chat.py

@@ -14,6 +14,7 @@ from agentbyte._optional import MissingExtraError
 from agentbyte.messages import AssistantMessage, Message, ToolCallRequest
 
 from agentbyte.component import Component
+from agentbyte.middleware.base import ClientMiddleware
 
 from .._retry_observability import build_retry_summary
 from ..base import (
@@ -70,8 +71,9 @@ class AzureOpenAIChatCompletionClient(
         initial_retry_delay: float = 1.0,
         max_retry_delay: float = 60.0,
         retry_policy: Optional[RetryPolicy] = None,
+        middlewares: Optional[List[ClientMiddleware]] = None,
     ):
-        super().__init__(model, client, config)
+        super().__init__(model, client, config, middlewares=middlewares)
         self.model_pricing = model_pricing
         self.pricing_model = pricing_model
         if retry_policy is not None:
@@ -109,6 +111,7 @@ class AzureOpenAIChatCompletionClient(
         initial_retry_delay: float = 1.0,
         max_retry_delay: float = 60.0,
         retry_policy: Optional[RetryPolicy] = None,
+        middlewares: Optional[List[ClientMiddleware]] = None,
     ) -> "AzureOpenAIChatCompletionClient":
         """Create a client using an Azure OpenAI API key."""
         from pydantic import ValidationError
@@ -159,6 +162,7 @@ class AzureOpenAIChatCompletionClient(
             initial_retry_delay=initial_retry_delay,
             max_retry_delay=max_retry_delay,
             retry_policy=retry_policy,
+            middlewares=middlewares,
         )
 
     @classmethod
@@ -175,6 +179,7 @@ class AzureOpenAIChatCompletionClient(
         initial_retry_delay: float = 1.0,
         max_retry_delay: float = 60.0,
         retry_policy: Optional[RetryPolicy] = None,
+        middlewares: Optional[List[ClientMiddleware]] = None,
     ) -> "AzureOpenAIChatCompletionClient":
         """Create a client using DefaultAzureCredential."""
         try:
@@ -226,6 +231,7 @@ class AzureOpenAIChatCompletionClient(
             initial_retry_delay=initial_retry_delay,
             max_retry_delay=max_retry_delay,
             retry_policy=retry_policy,
+            middlewares=middlewares,
         )
 
     @classmethod
@@ -243,6 +249,7 @@ class AzureOpenAIChatCompletionClient(
         initial_retry_delay: float = 1.0,
         max_retry_delay: float = 60.0,
         retry_policy: Optional[RetryPolicy] = None,
+        middlewares: Optional[List[ClientMiddleware]] = None,
     ) -> "AzureOpenAIChatCompletionClient":
         """Create a client using certificate-based Azure AD service principal auth."""
         from pydantic import ValidationError
@@ -303,6 +310,7 @@ class AzureOpenAIChatCompletionClient(
             initial_retry_delay=initial_retry_delay,
             max_retry_delay=max_retry_delay,
             retry_policy=retry_policy,
+            middlewares=middlewares,
         )
 
     @classmethod
@@ -319,6 +327,7 @@ class AzureOpenAIChatCompletionClient(
         initial_retry_delay: float = 1.0,
         max_retry_delay: float = 60.0,
         retry_policy: Optional[RetryPolicy] = None,
+        middlewares: Optional[List[ClientMiddleware]] = None,
     ) -> "AzureOpenAIChatCompletionClient":
         """Create a client using a pre-fetched Azure AD bearer token."""
         from pydantic import ValidationError
@@ -362,16 +371,18 @@ class AzureOpenAIChatCompletionClient(
             initial_retry_delay=initial_retry_delay,
             max_retry_delay=max_retry_delay,
             retry_policy=retry_policy,
+            middlewares=middlewares,
         )
 
-    async def create(
+    async def _execute_chat_operation(
         self,
-        messages: List[Message],
-        tools: Optional[List[Dict[str, Any]]] = None,
-        output_format: Optional[type] = None,
-        **kwargs: Any,
+        payload: Dict[str, Any],
     ) -> ChatCompletionResult:
         """Make a single Azure OpenAI API call."""
+        messages = payload["messages"]
+        tools = payload["tools"]
+        output_format = payload["output_format"]
+        kwargs = payload["kwargs"]
         retry_history: List = []
         try:
             result = await self._retry_with_backoff(

{agentbyte-0.52.0 → agentbyte-0.53.0}/src/agentbyte/llm/azure/embedding.py

@@ -10,6 +10,7 @@ from openai import AsyncAzureOpenAI
 
 from agentbyte._optional import MissingExtraError
 from agentbyte.component import Component
+from agentbyte.middleware.base import ClientMiddleware
 
 from .._retry_observability import build_retry_summary
 from ..base import AuthenticationError, InvalidRequestError, RateLimitError
@@ -53,8 +54,9 @@ class AzureOpenAIEmbeddingClient(
         initial_retry_delay: float = 1.0,
         max_retry_delay: float = 60.0,
         retry_policy: Optional[RetryPolicy] = None,
+        middlewares: Optional[List[ClientMiddleware]] = None,
     ):
-        super().__init__(model, client, config)
+        super().__init__(model, client, config, middlewares=middlewares)
         self.model_pricing = model_pricing
         self.pricing_model = pricing_model
         if retry_policy is not None:
@@ -93,6 +95,7 @@ class AzureOpenAIEmbeddingClient(
         initial_retry_delay: float = 1.0,
         max_retry_delay: float = 60.0,
         retry_policy: Optional[RetryPolicy] = None,
+        middlewares: Optional[List[ClientMiddleware]] = None,
     ) -> "AzureOpenAIEmbeddingClient":
         """Create client using Azure OpenAI API key authentication."""
         from pydantic import ValidationError
@@ -143,6 +146,7 @@ class AzureOpenAIEmbeddingClient(
             initial_retry_delay=initial_retry_delay,
             max_retry_delay=max_retry_delay,
             retry_policy=retry_policy,
+            middlewares=middlewares,
         )
 
     @classmethod
@@ -160,6 +164,7 @@ class AzureOpenAIEmbeddingClient(
         initial_retry_delay: float = 1.0,
         max_retry_delay: float = 60.0,
         retry_policy: Optional[RetryPolicy] = None,
+        middlewares: Optional[List[ClientMiddleware]] = None,
     ) -> "AzureOpenAIEmbeddingClient":
         """Create client using DefaultAzureCredential token provider."""
         try:
@@ -211,6 +216,7 @@ class AzureOpenAIEmbeddingClient(
             initial_retry_delay=initial_retry_delay,
             max_retry_delay=max_retry_delay,
             retry_policy=retry_policy,
+            middlewares=middlewares,
         )
 
     @classmethod
@@ -229,6 +235,7 @@ class AzureOpenAIEmbeddingClient(
         initial_retry_delay: float = 1.0,
         max_retry_delay: float = 60.0,
         retry_policy: Optional[RetryPolicy] = None,
+        middlewares: Optional[List[ClientMiddleware]] = None,
     ) -> "AzureOpenAIEmbeddingClient":
         """Create client using certificate-based Azure AD authentication."""
         from pydantic import ValidationError
@@ -289,6 +296,7 @@ class AzureOpenAIEmbeddingClient(
             initial_retry_delay=initial_retry_delay,
             max_retry_delay=max_retry_delay,
             retry_policy=retry_policy,
+            middlewares=middlewares,
         )
 
     @classmethod
@@ -306,6 +314,7 @@ class AzureOpenAIEmbeddingClient(
         initial_retry_delay: float = 1.0,
         max_retry_delay: float = 60.0,
         retry_policy: Optional[RetryPolicy] = None,
+        middlewares: Optional[List[ClientMiddleware]] = None,
     ) -> "AzureOpenAIEmbeddingClient":
         """Create client using a pre-fetched Azure AD bearer token."""
         from pydantic import ValidationError
@@ -349,6 +358,7 @@ class AzureOpenAIEmbeddingClient(
             initial_retry_delay=initial_retry_delay,
             max_retry_delay=max_retry_delay,
             retry_policy=retry_policy,
+            middlewares=middlewares,
         )
 
     async def create(self, input_text: str, **kwargs: Any) -> EmbeddingResult:

{agentbyte-0.52.0 → agentbyte-0.53.0}/src/agentbyte/llm/base.py

@@ -21,6 +21,7 @@ from pydantic import BaseModel
 
 from agentbyte.component import ComponentLoader
 from agentbyte.messages import AssistantMessage, Message, MultiModalMessage, ToolMessage
+from agentbyte.middleware.base import ClientMiddleware, MiddlewareChain, OperationType
 
 from .types import (
     ChatCompletionChunk,
@@ -171,6 +172,7 @@ class BaseChatCompletionClient(ABC, ComponentLoader):
         model: str,
         client: Any,
         config: Optional[Dict[str, Any]] = None,
+        middlewares: Optional[List[ClientMiddleware]] = None,
         **kwargs: Any,
     ):
         """
@@ -226,8 +228,26 @@ class BaseChatCompletionClient(ABC, ComponentLoader):
         self.config = config or {}
         # Additional kwargs override config
         self.config.update(kwargs)
+        for middleware in middlewares or []:
+            if not isinstance(middleware, ClientMiddleware):
+                raise TypeError(
+                    f"{type(middleware).__name__} is not a ClientMiddleware — only "
+                    "middleware that does not require AgentContext or conversational "
+                    "state (e.g. OTelMiddleware, LoggingMiddleware) can attach "
+                    "directly to a client."
+                )
+        self._middleware_chain = MiddlewareChain(list(middlewares or []))
+
+    @property
+    def middleware_types(self) -> frozenset:
+        """Concrete types of middleware attached to this client.
+
+        Public introspection surface for callers that need to check this
+        client's middleware without reaching into ``_middleware_chain``
+        directly (e.g. a wrapping ``Agent`` detecting duplicate instrumentation).
+        """
+        return frozenset(type(m) for m in self._middleware_chain.middlewares)
 
-    @abstractmethod
     async def create(
         self,
         messages: List[Message],
@@ -338,7 +358,23 @@ class BaseChatCompletionClient(ABC, ComponentLoader):
             print(f"Tokens: {result.usage.total_tokens}")
             ```
         """
-        pass
+        return await self._middleware_chain.execute(
+            operation=OperationType.MODEL_CALL.value,
+            agent_name=type(self).__name__,
+            agent_context=None,
+            data={
+                "messages": messages,
+                "tools": tools,
+                "output_format": output_format,
+                "kwargs": kwargs,
+            },
+            func=self._execute_chat_operation,
+            metadata={"model": self.model},
+        )
+
+    async def _execute_chat_operation(self, payload: Dict[str, Any]) -> ChatCompletionResult:
+        """Execute a non-streaming provider call after middleware processing."""
+        raise NotImplementedError
 
     @abstractmethod
     async def create_stream(
@@ -636,4 +672,3 @@ __all__ = [
     "AuthenticationError",
     "InvalidRequestError",
 ]
-

{agentbyte-0.52.0 → agentbyte-0.53.0}/src/agentbyte/llm/bedrock/anthropic_chat.py

@@ -1,11 +1,12 @@
 """Anthropic Claude chat client transported through AWS Bedrock."""
 
 import re
-from typing import Any, Dict, Optional
+from typing import Any, Dict, List, Optional
 
 from anthropic import AsyncAnthropicBedrock
 
 from agentbyte.component import Component
+from agentbyte.middleware.base import ClientMiddleware
 from agentbyte.llm.anthropic._conversion import AnthropicConversionMixin
 from agentbyte.llm.base import BaseChatCompletionClient, BaseChatCompletionClientConfig
 from agentbyte.llm.pricing import PricingRegistry
@@ -67,8 +68,9 @@ class AnthropicBedrockChatCompletionClient(
         initial_retry_delay: float = 1.0,
         max_retry_delay: float = 60.0,
         retry_policy: Optional[RetryPolicy] = None,
+        middlewares: Optional[List[ClientMiddleware]] = None,
     ):
-        super().__init__(model, client, config)
+        super().__init__(model, client, config, middlewares=middlewares)
         self.model_pricing = model_pricing
         if retry_policy is not None:
             self.retry_policy = retry_policy
@@ -103,6 +105,7 @@ class AnthropicBedrockChatCompletionClient(
         initial_retry_delay: float = 1.0,
         max_retry_delay: float = 60.0,
         retry_policy: Optional[RetryPolicy] = None,
+        middlewares: Optional[List[ClientMiddleware]] = None,
     ) -> "AnthropicBedrockChatCompletionClient":
         """Create a client from an explicit or env-backed AWS region/profile.
 
@@ -128,6 +131,7 @@ class AnthropicBedrockChatCompletionClient(
             initial_retry_delay=initial_retry_delay,
             max_retry_delay=max_retry_delay,
             retry_policy=retry_policy,
+            middlewares=middlewares,
         )
 
     def _estimate_cost(

{agentbyte-0.52.0 → agentbyte-0.53.0}/src/agentbyte/llm/bedrock/cohere_embedding.py

@@ -7,6 +7,7 @@ from typing import Any, Dict, List, Optional
 
 from agentbyte.component import Component
 from agentbyte.llm._retry_observability import build_retry_summary
+from agentbyte.middleware.base import ClientMiddleware
 from agentbyte.llm.base import AuthenticationError, InvalidRequestError, RateLimitError
 from agentbyte.llm.embeddings_base import BaseEmbeddingClient, BaseEmbeddingClientConfig
 from agentbyte.llm.pricing import PricingRegistry
@@ -49,8 +50,8 @@ class CohereBedrockEmbeddingClient(
                  model_pricing: Optional[ModelPricing] = None, max_retries: int = 3,
                  initial_retry_delay: float = 1.0, max_retry_delay: float = 60.0,
                  retry_policy: Optional[RetryPolicy] = None, region: str | None = None,
-                 profile: str | None = None):
-        super().__init__(model, client, config)
+                 profile: str | None = None, middlewares: Optional[List[ClientMiddleware]] = None):
+        super().__init__(model, client, config, middlewares=middlewares)
         self.model_pricing, self.region, self.profile = model_pricing, region, profile
         self.retry_policy = retry_policy or RetryPolicy(max_retries=max_retries, initial_retry_delay=initial_retry_delay, max_retry_delay=max_retry_delay)
 
@@ -58,11 +59,12 @@ class CohereBedrockEmbeddingClient(
     def from_region(cls, model: str, region: str | None = None, profile: str | None = None,
                     config: Optional[Dict[str, Any]] = None, model_pricing: Optional[ModelPricing] = None,
                     max_retries: int = 3, initial_retry_delay: float = 1.0, max_retry_delay: float = 60.0,
-                    retry_policy: Optional[RetryPolicy] = None) -> "CohereBedrockEmbeddingClient":
+                    retry_policy: Optional[RetryPolicy] = None,
+                    middlewares: Optional[List[ClientMiddleware]] = None) -> "CohereBedrockEmbeddingClient":
         settings = BedrockSettings()
         return cls(model, config=config, model_pricing=model_pricing, max_retries=max_retries,
                    initial_retry_delay=initial_retry_delay, max_retry_delay=max_retry_delay, retry_policy=retry_policy,
-                   region=region or settings.region, profile=profile or settings.profile)
+                   region=region or settings.region, profile=profile or settings.profile, middlewares=middlewares)
 
     async def create(self, input_text: str, **kwargs: Any) -> EmbeddingResult:
         return await self.create_batch([input_text], **kwargs)