---
id: 20260926-T0-06
title: "给LLM Agent装上「权限门」：按角色逐步开放工具，防止越权调用"
title_en: "Role-Scoped Access Control Limits Tool Misuse in LLM Agents"
url: https://ai.daily.yangsir.net/daily/20260926-T0-06
issue_date: 2026-09-26
publish_date: 2026-09-25T04:00:00.000Z
category: research
source_name: "arXiv cs.AI"
source_url: https://arxiv.org/abs/2609.28693
---

# 给LLM Agent装上「权限门」：按角色逐步开放工具，防止越权调用

一篇新论文提出用渐进式技能发现（Progressive Skill Discovery）作为工具使用型LLM智能体的访问控制机制。当智能体接入大量企业工具时，直接开放全部工具会导致上下文窗口膨胀、工具选择质量下降和严重的治理漏洞。该方法通过角色范围的能力交付实现结构性治理，让智能体根据角色逐步解锁可用工具，而非一次性暴露全部接口。这为企业安全部署LLM智能体提供了可操作的框架，开发者可以在不让智能体"看到"全部工具的前提下，按需分配权限，降低误用和越权风险。

## English Version

**Role-Scoped Access Control Limits Tool Misuse in LLM Agents**

A new paper proposes Progressive Skill Discovery as an access control mechanism for tool-using LLM agents. When agents are given access to every enterprise tool, it leads to oversized context windows, degraded tool selection, and governance vulnerabilities. The method implements structural governance through role-scoped capability delivery, letting agents progressively unlock tools based on their role rather than exposing all interfaces at once. This provides an actionable framework for enterprises deploying LLM agents safely — developers can assign permissions on demand without exposing the full toolset, reducing misuse and privilege escalation risks.

---

**来源**：[arXiv cs.AI](https://arxiv.org/abs/2609.28693)

**详情页**：https://ai.daily.yangsir.net/daily/20260926-T0-06

---

*智语观潮 · Daily — https://ai.daily.yangsir.net/llms.txt*