# Bolt 新智能体默认内置安全扫描

- 来源：Rohan Paul (@rohanpaul_ai)
- 发布时间：2026-07-31 23:57
- AIHOT 分数：39
- AIHOT 链接：https://aihot.virxact.com/items/cms94nydo04d8ro9k4okzveyk
- 原文链接：https://x.com/rohanpaul_ai/status/2083220481491304746

## AI 摘要

Bolt 发布新智能体，将安全工程师内置到每个项目中，默认开启且免费。该智能体在发布前对应用进行 6 类深度扫描：访问控制、信息泄露、认证、业务逻辑、不可信输入及配置与密钥，并在修复与部署之间显式加入构建验证，以应对 AI 加速开发带来的安全风险。

## 正文

Faster code generation also increases the rate at which insecure assumptions reach production.

But more scanning alone does not match the speed of an app builder.

@boltdotnew 's new agent closes that loop inside publishing.

The agent deep scans the app across 6 categories： access control， information disclosure， authentication， business logic， untrusted input， and configuration and secrets.

It explicitly includes build verification between remediation and deployment.

### 引用推文

> bolt.new：Today we're shipping a security engineer into every Bolt project. AI made building faster. It made attackers faster too. So we've made SoTA agentic security the...