Merge pull request #416 from agentuniverse-ai/Release_20250615

feat: Release version 0.0.18

Author: LandJerry

CHANGELOG.md

@@ -24,6 +24,27 @@ Note - Additional remarks regarding the version.
 ***************************************************
 
 # Version Update History
+## [0.0.18] - 2025-07-10
+### Added
+- Added observability capabilities for agent applications based on the OTel protocol
+Established observation standards for aU agent applications based on OTel (OpenTelemetry) protocol. Comprehensive metrics and traces (including token consumption, latency, success rate, etc.) are collected across key components such as agents, LLMs, and tools. The implementation supports integration with mainstream observability frameworks including SigNoz, Jaeger, and Prometheus, enabling full lifecycle observability for agents.
+
+- Added async Sink and Sender components for SLS Logger
+
+### Fixed
+- MCP service in Stdio mode now supports user-configurable environment parameter
+
+### Note
+- Third-party package dependency changes (Format: old version -> new version; single version means new)
+  - openai ("1.13.3" -> "1.55.3")
+  - opentelemetry-api ("^1.25.0")
+  - opentelemetry-sdk ("^1.25.0")
+  - opentelemetry-semantic-conventions (">=0.48b0")
+  - opentelemetry-exporter-otlp-proto-http ("^1.25.0")
+  - httpx ("0.27.2" -> ">=0.27.2")
+  - jsonlines ("^4.0.0")
+- Other code optimizations and documentation updates
+
 ## [0.0.17] - 2025-05-22
 ### Added
 - MCP Integration & Publishing Capabilities   

CHANGELOG_zh.md

@@ -24,6 +24,27 @@ Note - 对于版本的额外说明。
 ***************************************************
 
 # 版本更新记录
+## [0.0.18] - 2025-07-10
+### Added
+- 新增基于OTel协议的智能体应用可观测能力
+制定aU智能体应用的观测标准，基于OTel(OpenTelemetry)协议，对智能体、模型、工具等关键组件的token消耗、耗时、成功率等核心指标以及调用链（trace）进行全面采集与观测，并适配 SigNoz、Jaeger、Prometheus 等主流观测框架，从而实现对智能体全生命周期的可观测性支持。
+
+- 新增异步SLS日志Sink、Sender组件
+
+### Fixed
+- MCP服务调用Stdio模式下env参数支持用户配置透传
+
+### Note
+- 第三方包依赖变更（格式：原版本->新版本，只包含一个版本号为新引入）
+  - openai ("1.13.3" -> "1.55.3")
+  - opentelemetry-api ("^1.25.0")
+  - opentelemetry-sdk ("^1.25.0")
+  - opentelemetry-semantic-conventions (">=0.48b0")
+  - opentelemetry-exporter-otlp-proto-http ("^1.25.0")
+  - httpx ("0.27.2" -> ">=0.27.2")
+  - jsonlines ("^4.0.0")
+- 其他代码优化与文档更新
+
 ## [0.0.17] - 2025-05-22
 ### Added
 - 新增MCP接入与发布能力

README.md

@@ -5,7 +5,7 @@ Language version: [English](./README.md) | [中文](./README_zh.md) | [日本語
 ![](https://img.shields.io/badge/framework-agentUniverse-pink)
 ![](https://img.shields.io/badge/python-3.10%2B-blue?logo=Python)
 [![](https://img.shields.io/badge/%20license-Apache--2.0-yellow)](LICENSE)
-[![Static Badge](https://img.shields.io/badge/pypi-v0.0.17-blue?logo=pypi)](https://pypi.org/project/agentUniverse/)
+[![Static Badge](https://img.shields.io/badge/pypi-v0.0.18-blue?logo=pypi)](https://pypi.org/project/agentUniverse/)
 
 ![](docs/guidebook/_picture/logo_bar.jpg)
 ****************************************
@@ -89,6 +89,9 @@ You can learn how to create effective agent patterns into templates through the
 #### Using and Publishing MCP Servers
 You can learn how to quickly use or publish MCP servers in agentUniverse framework by referring to [How to Use MCP Servers](docs/guidebook/en/How-to/Use%20and%20Publish%20MCP%20Server/How_to_Use_MCP_Servers.md) and [How to Publish MCP Servers](docs/guidebook/en/How-to/Use%20and%20Publish%20MCP%20Server/How_to_Publish_MCP_Servers.md).
 
+#### Agent Application Observability
+agentUniverse provides a standardized observability framework for agent applications. Based on the OpenTelemetry protocol, it enables comprehensive data collection and monitoring of critical components such as agents, LLMs, and tools, facilitating full lifecycle tracking of agents. You can learn how to utilize these observability capabilities in the [Agent Application Observability](docs/guidebook/en/In-Depth_Guides/Tech_Capabilities/Observability/Observability_with_OTEL.md) section.
+
 #### Common Tips and Advanced Techniques
 You can learn more advanced techniques in the agent application building process through other documents in the [Get_Start](./docs/guidebook/en/Get_Start), such as how to add a memory module into the intelligent agent process and how to effectively manage prompts within the project.
 ****************************************

README_zh.md

@@ -5,7 +5,7 @@
 ![](https://img.shields.io/badge/framework-agentUniverse-pink)
 ![](https://img.shields.io/badge/python-3.10%2B-blue?logo=Python)
 [![](https://img.shields.io/badge/%20license-Apache--2.0-yellow)](LICENSE)
-[![Static Badge](https://img.shields.io/badge/pypi-v0.0.17-blue?logo=pypi)](https://pypi.org/project/agentUniverse/)
+[![Static Badge](https://img.shields.io/badge/pypi-v0.0.18-blue?logo=pypi)](https://pypi.org/project/agentUniverse/)
 
 ![](docs/guidebook/_picture/logo_bar.jpg)
 ****************************************
@@ -26,8 +26,8 @@
 
 The LLM model integration can be accomplished with simple configuration, currently agentUniverse supported models include:
 
-|-|供应商|                                                        模型                                                         |
-|:-----:|:--------:|:-----------------------------------------------------------------------------------------------------------------:|
+|-|供应商|                                                      模型                                                      |
+|:-----:|:--------:|:------------------------------------------------------------------------------------------------------------:|
 |<img src="https://github.com/user-attachments/assets/b7b0f2ce-3250-4008-b6d7-4712a983deb9" height="25">|Qwen| qwen3 Series（qwen3-235b-a22b、qwen3-32b、qwen3-30b-a3b, etc.） 、qwen2.5-72b-instruct、qwq-32b-preview、qwen-max、… |
 |<img src="https://github.com/user-attachments/assets/5a997feb-bef4-4e53-ac3e-d38221e5399c" height="25">|Deepseek|                            deepseek-r1、deepseek-v3、deepseek-r1-distill-qwen-32b、…                            |
 |<img src="https://github.com/user-attachments/assets/0b50e555-65e8-49b2-b725-f3f71ee7daed" height="25">|OpenAI|                                GPT-4o、GPT-4o mini、OpenAI o1、OpenAI o3-mini、…                                 |
@@ -98,6 +98,9 @@ pip install agentUniverse
 #### 使用与发布MCP服务
 您可以通过 [如何使用MCP服务](./docs/guidebook/zh/How-to/使用与发布MCP服务/如何使用MCP服务.md)、[如何发布MCP服务](./docs/guidebook/zh/How-to/使用与发布MCP服务/如何发布MCP服务.md) 章节掌握如何快速在智能体框架中使用或发布MCP服务。
 
+#### 智能体应用观测
+agentUniverse拥有标准的智能体应用观测标准，并基于OpenTelemetry协议对于智能体、模型、工具等重要组件进行全方位采集与观测，方便用户追踪智能体全生命周期，您可以在 [智能体应用可观测](./docs/guidebook/zh/In-Depth_Guides/技术组件/可观测/基于OTEL的可观测能力.md) 章节掌握如何使用观测能力。
+
 #### 常用使用技巧
 您可以通过 [开始章节](./docs/guidebook/zh/开始使用) 下的其他文档了解智能体应用构建过程中的其他进阶技巧，例如如何在智能体过程中加入记忆模块、如何有效的管理项目中的prompt等。
 

agentuniverse/agent/action/knowledge/doc_processor/token_text_splitter.py

@@ -27,9 +27,12 @@ class TokenTextSplitter(DocProcessor):
     @property
     def splitter(self) -> Splitter:
         if not self.__splitter:
-            self.__splitter = Splitter(separator=self.separator,
-                                       chunk_size=self.chunk_size,
-                                       chunk_overlap=self.chunk_overlap)
+            self.__splitter = Splitter(
+            encoding_name=self.encoding_name,
+            model_name=self.model_name,
+            chunk_size=self.chunk_size,
+            chunk_overlap=self.chunk_overlap
+        )
         return self.__splitter
 
     def _process_docs(self, origin_docs: List[Document], query: Query = None) -> \

agentuniverse/agent/agent.py

@@ -46,7 +46,7 @@
 from agentuniverse.base.util.logging.logging_util import LOGGER
 from agentuniverse.base.util.memory_util import generate_messages, get_memory_string
 from agentuniverse.base.util.system_util import process_dict_with_funcs, is_system_builtin
-from agentuniverse.base.util.tracing.au_trace_manager import AuTraceManager
+from agentuniverse.base.tracing.au_trace_manager import AuTraceManager
 from agentuniverse.llm.llm import LLM
 from agentuniverse.llm.llm_manager import LLMManager
 from agentuniverse.prompt.chat_prompt import ChatPrompt

agentuniverse/agent_serve/web/gunicorn_server.py

@@ -13,7 +13,7 @@
 from .post_fork_queue import POST_FORK_QUEUE
 from ...base.annotation.singleton import singleton
 from ...base.context.framework_context_manager import FrameworkContextManager
-from ...base.util.tracing.au_trace_manager import AuTraceManager
+from agentuniverse.base.tracing.au_trace_manager import AuTraceManager
 
 DEFAULT_GUNICORN_CONFIG = {
     'bind': '127.0.0.1:8888',

agentuniverse/agent_serve/web/request_task.py

@@ -25,7 +25,7 @@
 from .thread_with_result import ThreadWithReturnValue
 from agentuniverse.base.util.logging.logging_util import LOGGER
 from ...agent.output_object import OutputObject
-from agentuniverse.base.util.tracing.au_trace_manager import AuTraceManager
+from agentuniverse.base.tracing.au_trace_manager import AuTraceManager
 from agentuniverse.base.util.logging.general_logger import get_context_prefix
 from agentuniverse.base.util.logging.log_type_enum import LogTypeEnum
 

agentuniverse/agent_serve/web/thread_with_result.py

@@ -6,10 +6,11 @@
 # @Email   : fanen.lhy@antgroup.com
 # @FileName: thread_with_result.py
 
-import weakref
-from concurrent.futures.thread import ThreadPoolExecutor, _worker, \
-    _threads_queues
+from concurrent.futures import ThreadPoolExecutor, Future
 from threading import Thread
+from typing import Callable
+
+from opentelemetry import context as otel_context
 
 from agentuniverse.base.context.context_coordinator import ContextCoordinator
 
@@ -34,11 +35,14 @@ def run(self):
         """Run the target func and save result in _return."""
         if self.target is not None:
             # set the context values in the thread
-            ContextCoordinator.recover_context(self._context_pack)
+            otel_token = ContextCoordinator.recover_context(self._context_pack)
             try:
                 self._return = self.target(*self.args, **self.kwargs)
             except Exception as e:
                 self.error = e
+            finally:
+                if otel_token:
+                    otel_context.detach(otel_token)
 
     def result(self):
         """Wait for target func finished, then return the result or raise an
@@ -49,27 +53,58 @@ def result(self):
         return self._return
 
 
+class ContextAwareFuture(Future):
+    """A Future that can store context and error information."""
+
+    def __init__(self):
+        super().__init__()
+        self.error = None
+        self._context_pack = None
+
+
 class ThreadPoolExecutorWithReturnValue(ThreadPoolExecutor):
+    """A ThreadPoolExecutor that preserves context across threads."""
+
+    def submit(self, fn: Callable, *args, **kwargs) -> ContextAwareFuture:
+        """Submit a callable to be executed with context preservation."""
+        context_pack = ContextCoordinator.save_context()
+        future = ContextAwareFuture()
+        future._context_pack = context_pack
+
+        def context_wrapper():
+            otel_token = None
+            try:
+                otel_token = ContextCoordinator.recover_context(context_pack)
+
+                result = fn(*args, **kwargs)
+                future.set_result(result)
+
+            except Exception as e:
+                future.error = e
+                future.set_exception(e)
+
+            finally:
+                if otel_token:
+                    otel_context.detach(otel_token)
+
+        super().submit(context_wrapper)
+
+        return future
+
+    def map(self, func: Callable, *iterables, timeout=None, chunksize=1):
+        """Apply func to each element of iterables with context preservation."""
+        context_pack = ContextCoordinator.save_context()
+
+        def context_aware_func(*args):
+            otel_token = None
+            try:
+                otel_token = ContextCoordinator.recover_context(context_pack)
+                return func(*args)
+            except Exception as e:
+                raise e
+            finally:
+                if otel_token:
+                    otel_context.detach(otel_token)
 
-    def _adjust_thread_count(self):
-        # if idle threads are available, don't spin new threads
-        if self._idle_semaphore.acquire(timeout=0):
-            return
-
-        # When the executor gets lost, the weakref callback will wake up
-        # the worker threads.
-        def weakref_cb(_, q=self._work_queue):
-            q.put(None)
-
-        num_threads = len(self._threads)
-        if num_threads < self._max_workers:
-            thread_name = '%s_%d' % (self._thread_name_prefix or self,
-                                     num_threads)
-            t = ThreadWithReturnValue(name=thread_name, target=_worker,
-                                 args=(weakref.ref(self, weakref_cb),
-                                       self._work_queue,
-                                       self._initializer,
-                                       self._initargs))
-            t.start()
-            self._threads.add(t)
-            _threads_queues[t] = self._work_queue
+        return super().map(context_aware_func, *iterables,
+                           timeout=timeout, chunksize=chunksize)

agentuniverse/base/agentuniverse.py

@@ -33,7 +33,7 @@
 from agentuniverse.agent_serve.web.web_booster import ACTIVATE_OPTIONS
 from agentuniverse.agent_serve.web.post_fork_queue import POST_FORK_QUEUE
 from agentuniverse.agent_serve.web.web_util import FlaskServerManager
-
+from agentuniverse.base.tracing.otel.telemetry_manager import TelemetryManager
 
 @singleton
 class AgentUniverse(object):
@@ -93,6 +93,9 @@ def start(self, config_path: str = None, core_mode: bool = False):
             config_path)
         init_loggers(log_config_path)
 
+        # Init OTEL configs
+        TelemetryManager().init_from_config(configer.value.get('OTEL', {}))
+
         # init web request task database
         RequestLibrary(configer=configer)