Agent Hacks Agent: Autoresearch for Production-Agent Red-Teaming
Production LLM agents such as Claude Code and Codex operate over untrusted content, files, commands, and workspace state, making safety failures directly actionable. Red-teaming must therefore keep pace with evolving models and tools. Existing approaches mainly optimize attack success and preserve artifacts such as benchmarks, payloads, or attack programs, which record where attacks succeed but not the enabling conditions behind unsafe agent behavior. We study automated red-teaming for production LLM agents using one agentic research environment to discover reusable vulnerability knowledge abo
Lineage graph
Paper → model → repo connections mined from source citations (Tier-1 exact match).
Why these links exist
Every edge carries a method, confidence, and the source snippet that justified it — so bad links are debuggable.
- PossiblePossibly related (embedding) · 60%huhusmang/Awesome-LLMs-for-Vulnerability-Detection →
- PossiblePossibly related (embedding) · 60%votal-ai-hq/wb-red-team →
- PossiblePossibly related (embedding) · 59%Best models for generating red-team attacks? Also looking for public datasets [R] →
- PossiblePossibly related (embedding) · 59%agent-tools →
- PossiblePossibly related (embedding) · 58%TracecatHQ/tracecat →
- PossiblePossibly related (embedding) · 28%SWE-agent/SWE-agent →
“Possibly related via embedding similarity 0.56 (not asserted). Timestamp check: artifact after paper (+3d).”
- LinkedLinked via arxiv author · 85%Xutao Mao →
“Agent Hacks Agent: Autoresearch for Production-Agent Red-Teaming”
- LinkedLinked via arxiv author · 85%Xiang Zheng →
“Agent Hacks Agent: Autoresearch for Production-Agent Red-Teaming”