PhD student at Fudan University, visiting student at HKUST. I study how to break and defend LLM-based agents — tool-invocation security, post-alignment tampering, and endogenous safety mechanisms.
Research: LLM Security • Agent Safety • Kernel Fuzzing • AI-enabled Vulnerability Detection
| Project | Description |
|---|---|
| TIPExploit | Empirical risk assessment of tool-invocation prompts in LLM agentic systems (Under Review) |
| HoneyPot | Attention honeypot experiments and Phase 0 validation artifacts |
| syzkaller (fork) | Kernel fuzzer enhanced with LLM-assisted mutation for improved coverage |
| Project | Description | Stars |
|---|---|---|
| Agent-Your-Agent | Multi-agent orchestration framework — routes tasks to the best model (Claude/Deepseek/GPT) via file-system protocol |  |
| SoulByte | Transform WeChat chat records into AI training datasets and personal knowledge bases |  |
| NexusAI-Hub | Unified platform for managing multi-provider AI models with OpenAI-compatible APIs |  |
| llm-roundtable | Multi-LLM structured debate platform — moderator/expert/critic agent roles |  |
| mcp_based_subsystem | MCP-based subsystem for AI agent coordination |  |
| Project | Description |
|---|---|
| mutilated_text_recognition | Deep learning model for recognizing mutilated/damaged text |
| Precious-Metals-Investment-Analysis | Quantitative analysis and investment strategy for precious metals |
| CaMeL-docs | Documentation site for the CaMeL ecosystem |
