English(EN) Stop trusting the agent: bind tool-call approvals to the exact call

代理系统通过签名审批保护工具调用

作者 PulseAugur 编辑部 · [1 个来源] · 2026-06-17 15:11

一种保护代理系统的新方法提出用加密签名的令牌替换简单的布尔审批标志。此方法将审批直接绑定到特定的工具调用，从而降低了提示注入、重放攻击和参数漂移的风险。通过对规范化参数、主体和到期时间使用 HMAC 签名，系统确保只授权特定的、预期的操作，防止未经授权的修改或重用审批。 AI

影响通过加密签名的审批防止提示注入和未经授权的操作，从而增强了 AI 代理的安全性。

排序理由该项目提出了一个针对代理系统安全问题的创新技术解决方案，包括代码示例。[lever_c_demoted from research: ic=1 ai=1.0]

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 0 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 该项目提出了一个针对代理系统安全问题的创新技术解决方案，包括代码示例。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: safety, infra; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: 107 days old; Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.

完整方法见我们的编辑标准。

dev.to — LLM tag TIER_1 English(EN) · Whatsonyourmind · 2026-06-17 15:11

停止信任代理：将工具调用批准绑定到确切的调用

<p>Agentic systems gate dangerous tool calls — file writes, money movement, deploys — behind an "approval": a human-in-the-loop click, or a policy check. Look at how that approval is usually represented and you'll often find a boolean sitting in the run/session state: <code>appro…