agentbyte 0.51.0__tar.gz → 0.53.0__tar.gz

{agentbyte-0.51.0 → agentbyte-0.53.0}/CHANGELOG.md

@@ -6,6 +6,80 @@ 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
+
+- `AnthropicChatCompletionClient`: direct Anthropic Claude chat client (`agentbyte[anthropic]`),
+  matching the existing `OpenAIChatCompletionClient` construction/DI/retry/pricing/error-mapping
+  contract (`from_api_key()`, `_to_config`/`_from_config` without persisting secrets, tool-use,
+  streaming).
+- `AnthropicBedrockChatCompletionClient`: the same Claude Messages API client transported through
+  AWS Bedrock (`agentbyte[bedrock]`) via `AsyncAnthropicBedrock` (SigV4, boto3 credential chain),
+  constructed with `from_region()`. Shares conversion/parsing logic with the direct client through
+  `AnthropicConversionMixin`.
+- `CohereBedrockEmbeddingClient`: Cohere Embed v3 (English/Multilingual) and v4 embeddings via AWS
+  Bedrock Runtime (`agentbyte[aws]`, `aioboto3`), with `CohereInputType`/`CohereTruncate` options
+  and the same injected-or-fresh-client pattern as `dataset/s3.py`'s `s3_client()`.
+- `build_chat_client` auto-detects `anthropic` (`ANTHROPIC_API_KEY`) and `anthropic-bedrock`
+  (`AWS_REGION` + `ANTHROPIC_BEDROCK_MODEL`) alongside the existing OpenAI/Azure detection.
+- Pricing for Claude 5.5 (`claude-opus-5-5`, `claude-haiku-5-5`), Cohere Embed v3 (English/
+  Multilingual, estimate), and Cohere Embed v4 (verified against AWS Bedrock's own pricing page).
+
+### Fixed
+
+- `AnthropicBedrockChatCompletionClient` was built on `AsyncAnthropicBedrockMantle`, a Bedrock
+  access path authenticated by a bearer token (`AWS_BEARER_TOKEN_BEDROCK`), not AWS IAM
+  credentials — every call 404'd with a misleading "model does not exist" error against a real
+  account. Caught by live smoke-testing against real Bedrock, invisible to the mocked unit tests.
+  Swapped to `AsyncAnthropicBedrock`, the real SigV4/boto3-credential-chain transport.
+- The Bedrock Claude pricing normalizer didn't strip cross-region inference-profile prefixes
+  (`eu.`/`global.`, required for on-demand invoke on most accounts), silently falling back to
+  default pricing for real Bedrock calls.
+- `CohereBedrockEmbeddingClient` read billed tokens from a response-body field that doesn't exist
+  on real Bedrock responses, so `tokens_input`/`cost_estimate` were silently always zero despite
+  successful calls. Now reads the real `x-amzn-bedrock-input-token-count` response header.
+
 ## [0.51.0] - 2026-10-09
 
 ### Added

{agentbyte-0.51.0 → agentbyte-0.53.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.5
 Name: agentbyte
-Version: 0.51.0
+Version: 0.53.0
 Summary: A toolkit for designing multiagent systems
 Author-email: MrDataPsycho <mr.data.psycho@gmail.com>
 License-Expression: LicenseRef-Proprietary
@@ -17,6 +17,8 @@ Requires-Dist: uvicorn[standard]>=0.44.0; extra == 'a2a'
 Provides-Extra: all
 Requires-Dist: aioboto3>=15.5.0; extra == 'all'
 Requires-Dist: aiosqlite>=0.22.1; extra == 'all'
+Requires-Dist: anthropic>=1.13.0; extra == 'all'
+Requires-Dist: anthropic[bedrock]>=1.13.0; extra == 'all'
 Requires-Dist: arxiv>=2.1; extra == 'all'
 Requires-Dist: asyncpg>=0.31.0; extra == 'all'
 Requires-Dist: azure-identity>=1.25.1; extra == 'all'
@@ -35,11 +37,15 @@ Requires-Dist: sqlmodel>=0.0.38; extra == 'all'
 Requires-Dist: tiktoken>=0.9.0; extra == 'all'
 Requires-Dist: uvicorn[standard]>=0.44.0; extra == 'all'
 Requires-Dist: youtube-transcript-api>=0.6; extra == 'all'
+Provides-Extra: anthropic
+Requires-Dist: anthropic>=1.13.0; extra == 'anthropic'
 Provides-Extra: aws
 Requires-Dist: aioboto3>=15.5.0; extra == 'aws'
 Provides-Extra: azureopenai
 Requires-Dist: azure-identity>=1.25.1; extra == 'azureopenai'
 Requires-Dist: openai>=1.107.1; extra == 'azureopenai'
+Provides-Extra: bedrock
+Requires-Dist: anthropic[bedrock]>=1.13.0; extra == 'bedrock'
 Provides-Extra: gunicorn
 Requires-Dist: fastapi>=0.135.2; extra == 'gunicorn'
 Requires-Dist: gunicorn>=23.0.0; extra == 'gunicorn'
@@ -84,7 +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
 
@@ -252,6 +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
@@ -333,7 +341,9 @@ uv add "agentbyte[openai,sqlite]"
 | --- | --- | --- |
 | `openai` | `openai` | `OpenAIChatCompletionClient`, `OpenAIEmbeddingClient` |
 | `azureopenai` | `openai`, `azure-identity` | Azure OpenAI clients and Entra ID credentials |
-| `aws` | `aioboto3` | S3 dataset sources, writers and publishers |
+| `anthropic` | `anthropic` | Direct Anthropic Claude chat client |
+| `bedrock` | `anthropic[bedrock]` | Anthropic Claude chat through AWS Bedrock |
+| `aws` | `aioboto3` | S3 dataset I/O and Cohere Bedrock embedding client |
 | `otel` | OpenTelemetry API, SDK, OTLP exporter | OpenTelemetry tracing and metrics |
 | `webui` | `fastapi`, `uvicorn` | `agentbyte webui` |
 | `gunicorn` | `webui` + `gunicorn` | Production single-worker serving (`gunicorn -k uvicorn.workers.UvicornWorker -w 1`) |

{agentbyte-0.51.0 → 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
@@ -251,7 +253,9 @@ uv add "agentbyte[openai,sqlite]"
 | --- | --- | --- |
 | `openai` | `openai` | `OpenAIChatCompletionClient`, `OpenAIEmbeddingClient` |
 | `azureopenai` | `openai`, `azure-identity` | Azure OpenAI clients and Entra ID credentials |
-| `aws` | `aioboto3` | S3 dataset sources, writers and publishers |
+| `anthropic` | `anthropic` | Direct Anthropic Claude chat client |
+| `bedrock` | `anthropic[bedrock]` | Anthropic Claude chat through AWS Bedrock |
+| `aws` | `aioboto3` | S3 dataset I/O and Cohere Bedrock embedding client |
 | `otel` | OpenTelemetry API, SDK, OTLP exporter | OpenTelemetry tracing and metrics |
 | `webui` | `fastapi`, `uvicorn` | `agentbyte webui` |
 | `gunicorn` | `webui` + `gunicorn` | Production single-worker serving (`gunicorn -k uvicorn.workers.UvicornWorker -w 1`) |

{agentbyte-0.51.0 → agentbyte-0.53.0}/pyproject.toml

@@ -24,6 +24,16 @@ azureopenai = [
     "openai>=1.107.1",
     "azure-identity>=1.25.1",
 ]
+anthropic = [
+    "anthropic>=1.13.0",
+]
+# anthropic[bedrock] does its own AWS SigV4 signing via AnthropicBedrock*; deliberately
+# not named "aws" (that extra is aioboto3, which this one does not need).
+bedrock = [
+    "anthropic[bedrock]>=1.13.0",
+]
+# aioboto3 for AWS I/O: S3 dataset access (dataset/s3.py) and Bedrock Runtime
+# (llm/bedrock/, e.g. CohereBedrockEmbeddingClient).
 aws = [
     "aioboto3>=15.5.0",
 ]
@@ -76,7 +86,7 @@ a2a = [
     "agentbyte[webui]",
 ]
 all = [
-    "agentbyte[openai,azureopenai,aws,otel,webui,gunicorn,sql,tiktoken,optim,viz,research]",
+    "agentbyte[openai,azureopenai,anthropic,bedrock,aws,otel,webui,gunicorn,sql,tiktoken,optim,viz,research]",
 ]
 
 

agentbyte-0.53.0/src/agentbyte/__about__.py

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

{agentbyte-0.51.0 → 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-0.51.0 → 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-0.51.0 → 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-0.51.0 → 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",
@@ -69,6 +73,8 @@ __all__ = [
     "AzureOpenAIChatCompletionClientConfig",
     "AzureOpenAIEmbeddingClient",
     "AzureOpenAIEmbeddingClientConfig",
+    "AnthropicChatCompletionClient",
+    "AnthropicChatCompletionClientConfig",
     # Authentication utilities
     "get_default_token_provider",
     "get_certificate_token_provider",
@@ -77,6 +83,12 @@ __all__ = [
     "OpenAISettings",
     "AzureOpenAISettings",
     "AzureServicePrincipalSettings",
+    "AnthropicSettings",
+    "AnthropicBedrockChatCompletionClient",
+    "AnthropicBedrockChatCompletionClientConfig",
+    "CohereBedrockEmbeddingClient",
+    "CohereBedrockEmbeddingClientConfig",
+    "BedrockSettings",
     # Retry policy
     "RetryPolicy",
     # Exception types
@@ -123,9 +135,22 @@ if TYPE_CHECKING:  # editors and type checkers see the provider classes eagerly
         OpenAIEmbeddingClientConfig,
         OpenAISettings,
     )
+    from .anthropic import (
+        AnthropicChatCompletionClient,
+        AnthropicChatCompletionClientConfig,
+        AnthropicSettings,
+    )
+    from .bedrock import (
+        AnthropicBedrockChatCompletionClient,
+        AnthropicBedrockChatCompletionClientConfig,
+        BedrockSettings,
+        CohereBedrockEmbeddingClient,
+        CohereBedrockEmbeddingClientConfig,
+    )
 
-# Provider clients need optional extras (``openai``, ``azureopenai``); they are
-# imported on first access so ``import agentbyte`` works with core dependencies only.
+# Provider clients need optional extras (``openai``, ``azureopenai``, ``anthropic``);
+# they are imported on first access so ``import agentbyte`` works with core
+# dependencies only.
 _PROVIDER_EXPORTS = {
     "OpenAIChatCompletionClient": ".openai",
     "OpenAIChatCompletionClientConfig": ".openai",
@@ -141,6 +166,25 @@ _PROVIDER_EXPORTS = {
     "create_token_provider_from_string": ".azure",
     "AzureOpenAISettings": ".azure",
     "AzureServicePrincipalSettings": ".azure",
+    "AnthropicChatCompletionClient": ".anthropic",
+    "AnthropicChatCompletionClientConfig": ".anthropic",
+    "AnthropicSettings": ".anthropic",
+    "AnthropicBedrockChatCompletionClient": ".bedrock",
+    "AnthropicBedrockChatCompletionClientConfig": ".bedrock",
+    "CohereBedrockEmbeddingClient": ".bedrock",
+    "CohereBedrockEmbeddingClientConfig": ".bedrock",
+    "BedrockSettings": ".bedrock",
+}
+
+_PROVIDER_EXTRAS = {
+    ".openai": ("openai", "openai"),
+    ".azure": ("openai", "azureopenai"),
+    ".anthropic": ("anthropic", "anthropic"),
+    # `.bedrock` is intentionally absent: its two clients need different extras
+    # (`anthropic[bedrock]` vs. `aioboto3`/`aws`), so `llm/bedrock/__init__.py`
+    # does its own fine-grained, per-symbol `require()` check instead -- a
+    # single module-level check here would wrongly demand both extras for
+    # either client.
 }
 
 
@@ -150,9 +194,12 @@ def __getattr__(name: str):
         raise AttributeError(f"module {__name__!r} has no attribute {name!r}")
     from importlib import import_module
 
-    from agentbyte._optional import require
+    extra_spec = _PROVIDER_EXTRAS.get(module_name)
+    if extra_spec is not None:
+        from agentbyte._optional import require
 
-    require("openai", extra="openai" if module_name == ".openai" else "azureopenai", feature=name)
+        required_module, extra = extra_spec
+        require(required_module, extra=extra, feature=name)
     value = getattr(import_module(module_name, __name__), name)
     globals()[name] = value
     return value

agentbyte-0.53.0/src/agentbyte/llm/anthropic/__init__.py

@@ -0,0 +1,10 @@
+"""Anthropic provider package."""
+
+from .chat import AnthropicChatCompletionClient, AnthropicChatCompletionClientConfig
+from .settings import AnthropicSettings
+
+__all__ = [
+    "AnthropicChatCompletionClient",
+    "AnthropicChatCompletionClientConfig",
+    "AnthropicSettings",
+]