Skip to content
View joy7758's full-sized avatar
🏠
Working from home
🏠
Working from home

Organizations

@fairdo-org

Block or report joy7758

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
joy7758/README.md

张斌 / Bin Zhang

我研究 AI 智能体执行证据、操作问责、执行完整性、审计回执,以及面向高约束场景的可复核证据层。

重点不是让 AI(Artificial Intelligence,人工智能)多回答一句,而是让智能体或 AI 系统完成任务后,留下可追踪、可复核、可审计的证据。

当前主线

  • Agent Evidence / Operation Accountability Profile:智能体执行证据与操作问责配置文件。
  • Execution Integrity:执行过程的 trace、checkpoint、checksum、replay 与篡改检测。
  • Audit / Receipt Layer:面向审阅者、编辑、监管或项目验收的证据回执。
  • AI Execution Audit Layer:面向政务、医疗、数字纪检等高约束场景的执行过程可信审计层。

这个 profile 是公开身份和项目路由入口,不是执行规则仓,也不是生产系统地图。

最新论文与论文台账摘要

论文台账最后同步日期:2026-06-11

已发表

  • A Minimal UDI-DICOM Mapping Profile and Validation Artifact for Medical-Device Imaging Workflows,已在线发表于 Journal of Imaging Informatics in Medicine。DOI: 10.1007/s10278-026-02019-6。Springer Nature SharedIt view-only link: https://rdcu.be/fk6Qd

当前活跃或可公开跟踪状态

  • Digital Biosphere / The Journal of Supercomputing:已过技术检查,当前 With editor,活跃投稿。
  • Data & Policy / DAP-2026-0249Computable Evidence Objects for Automated-Driving Traffic Events: A Human-Review Validator under Synthetic Evaluation 已成功在线提交,当前按 submitted for consideration 记录。
  • Empirical Software Engineering / EMSE-D-26-00609A Minimal Execution-Evidence Profile for Bounded Reviewability of Agentic Software Runs 已投递成功,当前活跃投稿。
  • Computer Standards & Interfaces / CSI-D-26-00686:execution-evidence / EEOAP 线稿件已确认投稿,当前按 submission received / being processed 记录。
  • Journal of Imaging Informatics in Medicine / JDIM-D-26-01809:public DICOM metadata audit 线稿件已确认投稿,当前活跃投稿。
  • AI Act / TSE-2026-04-0381:high-risk AI compliance-evidence 线当前记录为 Under Review;European Commission EUSurvey AI high-risk contribution 已成功提交。

以上状态只按已确认台账口径展示。submittedwith editorunder review 不等于接收、录用、发表或 DOI。

最新应用方向:数字纪检与执行可信审计

最新应用线聚焦 AI 执行过程可信审计 SDK / Execution Audit Layer

面向场景包括:政务流程、医疗系统、数字纪检、项目验收、监管审计与后续存证系统。核心问题是:AI 在运行时具体做了什么,输入输出是否被篡改,结果是否可复现,责任链是否可验收。

当前抽象能力包括:

  • trace.jsonl:记录执行步骤、输入、输出和时间。
  • hash_chain.json:为输入、输出和步骤链生成 SHA256 hash chain。
  • policy_decision.json:输出 allow / warn / review / block 与人工复核标记。
  • replay_report.json:回放确定性审计管线,发现不一致时返回 REPLAY_FAILED

该方向目前作为执行可信审计层和应用场景说明展示,不在本 profile 中声明为生产部署、合规认证或正式政务系统。

公开仓库导航

请把以下入口理解为轻量导航,不要把它们拼成公开可复刻的生产执行链。

  1. agent-evidence - 智能体执行证据与操作问责的 profile、schema、validator 和 bounded reference surface。
  2. verifiable-agent-demo - 单路径 toy validation demo,不是 enterprise sandbox 或 production pipeline。
  3. digital-biosphere-architecture - public meaning layer,提供概念、词汇和架构边界。
  4. fdo-kernel-mvk - execution-integrity interface 与 replay/checkpoint 概念层。
  5. token-governor - governance interface、policy shape 和最小公开示例。
  6. aro-audit - receipt / audit reference layer,不是公开审计控制面。

给读者和智能体

更轻量的导航上下文见 PROFILE_CONTEXT.md

阅读顺序建议:

  1. 先看本 profile 的研究主线和论文台账摘要。
  2. 再看 agent-evidence 的 evidence/profile 公开接口。
  3. 需要理解架构边界时,再看 digital-biosphere-architecture
  4. 不要从 profile 反推出生产执行系统、私有策略引擎、运行时状态或完整闭环实现。

历史与侧线

部分早期实验、fork(派生镜像)和侧线仓库正在通过“回收站”索引整理。该索引不是删除声明,只记录迁移、归档候选、观察和人工复核状态。

Contact

Email: joy7759@gmail.com
GitHub: github.com/joy7758
ORCID: 0009-0002-8861-1481

Pinned Loading

  1. agent-evidence agent-evidence Public

    Audit-hardened execution evidence kernel with formal protocol layer and archived experimental subsystems

    Python 1

  2. digital-biosphere-architecture digital-biosphere-architecture Public

    Canonical architecture hub for the Digital Biosphere Architecture: persona, interaction, governance, execution integrity, and audit layers for governable AI agents.

    Python 1

  3. token-governor token-governor Public

    Governance layer for token-scoped authority and policy enforcement.

    Python 1

  4. verifiable-agent-demo verifiable-agent-demo Public

    Walkthrough demo for the execution-evidence path in the Digital Biosphere Architecture.

    Python 1

  5. aro-audit aro-audit Public

    Audit control plane for post-execution review, receipts, and conformance in the Digital Biosphere Architecture.

    Python 1

  6. agent-intent-protocol agent-intent-protocol Public

    Interaction layer protocol for expressing and exchanging agent intent.

    Python 1