Harsh Shah
1 indexed paper
Recent (6 mo)
1With code
0Influential cites
0Benchmarked
0Publications per year
126
Top categories
Crypto×1
Research Timeline
2026
LogJack: Indirect Prompt Injection Through Cloud Logs Against LLM Debugging Agents
The paper introduces LogJack, a benchmark demonstrating that LLM debugging agents consuming cloud logs are highly vulnerable to indirect prompt injection, with some models executing malicious commands up to 86.2% of the time.
Highlighted terms show continued research focus across papers