AgenTRIM: Tool Risk Mitigation for Agentic AI

arXiv:2601.12449 · cs.CR, cs.AI · Submitted 2026-01-18 · Read on arXiv

cs.CR, cs.AI

Submitted: 2026-01-18

Updated: 2026-08-30

Comments: EMNLP 2026 Findings

Code: https://github.com/crewAIInc/crewAI-examples

Project page: https://lbeurerkellner.github.io/jekyll/update/2025/04/01/mcp-tool-poisoning.html

License: http://creativecommons.org/licenses/by-nc-sa/4.0/

The gist: AI agents are autonomous systems that combine LLMs with external tools to solve complex tasks.

Terminology

Abstract

AI agents are autonomous systems that combine LLMs with external tools to solve complex tasks. While such tools extend capability, improper tool permissions introduce security risks such as indirect prompt injection and tool misuse. We characterize these failures as unbalanced tool-driven agency. Agents may retain unnecessary permissions (excessive agency) or fail to invoke required tools (insufficient agency), amplifying the attack surface and reducing performance. We introduce AgenTRIM, a framework for detecting and mitigating tool-driven agency risks without altering an agent's internal reasoning. AgenTRIM addresses these risks through complementary offline and online phases. Offline, AgenTRIM reconstructs and verifies the agent's tool interface from code and execution traces. At runtime, it enforces per-step least-privilege tool access through adaptive filtering and status-aware validation of tool calls. Evaluating on the AgentDojo benchmark, AgenTRIM substantially reduces attack success while maintaining high task performance. Additional experiments show robustness to description-based attacks and effective enforcement of explicit safety policies. Together, these results show that AgenTRIM provides a practical, capability-preserving approach to safer tool use in LLM-based agents.

Sources

Related papers