mocX
中文

About

Beijing University of Posts and Telecommunications · Beijing

I publish code on GitHub as Zhong0118 and keep technical notes and working logs on the MocLab blog. This page only lists what can be checked against public repositories; degree, advisor, publications and experience stay hidden until the author confirms them.

Recent public projects cover three directions: supervising LLM agents at tool-call time, multi-agent organisations that change based on execution evidence, and detecting and attributing AI-generated content. There are also a few desktop and agent tools I use myself.

How I work

  1. Define the problem

    Start by stating the risk or collaboration problem. AgentGuard-Chain, for example, puts the security boundary after a tool call is generated and before it runs.

    AgentGuard-Chain

  2. Build a runnable prototype

    Turn the idea into the smallest system that runs: a security gateway, a strategy evolution loop, a memory store with provenance.

    AgentGuard-ChainEvoTeamMARS

  3. Check it with evidence

    Use scripted cases, unit tests or validation gates, and state the current status and limits in the README.

    AgentGuard-ChainEvoTeam

  4. Write it down in public

    Code goes to GitHub; notes on process and tools go to the MocLab blog.

Research directions

Three directions covered by the public projects, each linked to repositories you can inspect.

  • 01

    Agent security

    How can tool calls generated by an LLM agent be checked for privilege abuse and multi-step risk chains before they actually run?

    A security gateway between tool-call generation and execution audits, scores, blocks and logs file, command, API and outbound actions.

    Related projects

  • 02

    Multi-agent collaboration

    Can the roles and collaboration strategy of a multi-agent team change only after cross-task evidence and validation support it?

    EvoTeam constrains evolution with versioned strategies and a validation gate; MARS tries to turn team discussions into memory with provenance.

    Related projects

  • 03

    AIGC detection and attribution

    How can we tell whether content was generated by a model, and which model it came from?

    The aigc_detector repository is split into detection, provenance and ingestion modules; only its description is public so far, with no published results.

    Related projects

Skills

Only technologies used in public projects, without proficiency levels. Select one to see where it is used.

Languages

Front end & desktop

Back end & data

Agents & prototypes

Evidence

Python5 projects

Filter the project directory by “Python”

Education

No public entries listed

Research and work experience

No public entries listed

Awards & honors

No public entries listed

Contact

For now, reach me through GitHub or the MocLab blog.

Recent posts on MocLab

From the MocLab RSS feed, fetched 28 Oct 2026