智能体行为审计与存证数据集

本数据集聚焦AI智能体运行过程中的行为审计与存证，覆盖对话智能体、自动化流程智能体、决策支持智能体、工业控制智能体等类型。数据内容包括智能体元数据、行为事件（认证、数据访问、API调用、状态变更、通信等）、审计日志（含哈希链）、存证凭证（哈希值、数字签名、区块链存证模拟）及完整性验证结果。适用于审计系统开发、篡改检测、区块链存证验证、异常行为识别及合规审查等场景。This dataset focuses on behavior auditing and evidence preservation during the operation of AI Agents, covering multiple types such as conversational agents, automated workflow agents, decision support agents, and industrial control agents. The data content includes agent metadata, behavioral events (e.g., authentication, data access, API calls, state changes, communications, etc.), audit logs with built-in hash chains, evidence preservation credentials (including hash values, digital signatures, simulated blockchain-based evidence preservation), and integrity verification results. This dataset is applicable to scenarios including audit system development, tamper detection, blockchain-based evidence preservation verification, abnormal behavior recognition, and compliance review.

搜集汇总

背景与挑战

背景概述

该数据集专注于AI智能体运行过程中的行为审计与存证，涵盖对话、自动化流程、决策支持及工业控制等多种智能体类型。数据包含智能体元数据、行为事件、审计日志、存证凭证及验证结果，适用于审计系统开发、篡改检测、区块链存证验证、异常行为识别和合规审查等应用场景。