Securing GenAI Multi-Agent Systems Against Tool Squatting: A Zero Trust Registry-Based Approach
Vineeth Sai Narajala 1, Ken Huang 2, Idan Habler 3 · 2025
Read Securing GenAI Multi-Agent Systems Against Tool Squatting: A Zero Trust Registry-Based Approach by Vineeth Sai Narajala 1, Ken Huang 2, Idan Habler 3 (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2504-19951.txt.
Description
Vineeth Narajala, Ken Huang, and Idan Habler guard AI agent toolboxes against impostors in a 2025 paper. Multi-agent systems discover outside tools through shared protocols. A squatter can register a fake tool with a trusted name and intercept the work.
Their answer is a zero-trust registry. Administrators control registration; agents discover tools only through the registry. Fine-grained policies gate which agent may call which tool. Trust scores shift with version history and known flaws. Credentials are issued just in time, never stored with the agent. The design targets the common squatting routes while keeping the flexibility that makes agent systems useful. A foundation for production use is the stated goal.
Catalog record
| CREATOR | Vineeth Sai Narajala 1, Ken Huang 2, Idan Habler 3 |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 35 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2504-19951