3 个仓库
Interactive tools for visualizing and debugging the execution flow of agentic logic graphs.
Distinct from Interactive Graph Visualizers: Specializes general interactive graph visualizers [f3_mt1] for the purpose of debugging agentic workflow execution.
Explore 3 awesome GitHub repositories matching data & databases · Agent Graph Debuggers. Refine with filters or upvote what's useful.
AgentOps 是一个可观测性平台和开发者工具包,用于监控由大语言模型驱动的自主智能体的执行、性能和可靠性。它作为一个用于跟踪 AI 智能体行为、调试复杂工作流和基准测试模型性能的系统。 该平台的特点是能够通过执行路径图和会话回放来可视化多智能体工作流。它提供了用于计算跨不同语言模型提供商的财务支出的特定工具,并为需要在私有硬件或云端完全控制数据的用户提供自托管的可观测性栈。 该系统涵盖了一系列广泛的功能,包括智能体故障检测、工具使用分析以及通过事件标记跟踪自定义性能指标。它与 AI 框架集成以捕获遥测和性能数据。
Visualizes complex agent interactions by mapping causal relationships between inputs, outputs, and tool calls.
AdalFlow 是一个自主 AI 代理框架和 LLM 应用库,旨在构建模块化工作流。它作为一个模型无关的接口和 RAG 流水线编排器,允许用户开发 ReAct 代理,利用迭代推理和外部工具执行来解决复杂任务。 该项目通过一个提示词优化系统脱颖而出,该系统使用文本梯度下降自动优化提示词模板和少样本示例。它将模型反馈视为可微分信号,实现了一种 LLM 反向传播形式,从而根据评估指标迭代提高输出质量。 该框架涵盖了广泛的功能面,包括带有语义向量搜索和重排序的检索增强生成、用于可观测性的基于跨度的执行追踪,以及模式驱动的结构化解析。它为众多专有和开源模型提供商提供了统一的通信层,并支持将 Python 函数转换为标准化的工具接口。 该系统使用 Python 实现,并与 MLflow 集成以进行工作流跟踪和分析。
Provides interactive HTML and subgraph diagrams to visualize and debug the flow of agentic logic graphs.
This project provides a translation layer and set of adapters designed to bridge AI agents with the Model Context Protocol. It functions as an integration layer that allows agents to operate as protocol-compliant servers and enables the conversion of protocol-based tools into formats compatible with agent frameworks and logic graphs. The adapters facilitate tool interoperability by wrapping external protocol tools for use within agent workflows and exposing internal agent capabilities to any client implementing the Model Context Protocol. This creates a communication bridge that supports inte
Provides a specialized interface to visualize and interact with running logic graphs for testing and debugging.