warnStr = BlueBold + "%s\n" + Reset + Magenta + "[warn] " + Reset
		errStr = Magenta + "%s\n" + Reset + Red + "[error] " + Reset
		traceStr = Green + "%s\n" + Reset + Yellow + "[%.3fms] " + BlueBold + "[rows:%v]" + Reset + " %s"
		traceWarnStr = Green + "%s " + Yellow + "%s\n" + Reset + RedBold + "[%.3fms] " + Yellow + "[rows:%v]" + Magenta + " %s" + Reset

Imagina que construyes una forma de ejecutar un agente de IA local.

Provee un API en

```
http://localhost:8000/v1/agents/multivac
```

También hay un frontend en

```
http://localhost:8000
```

/// tip | Consejo

Ten en cuenta que ambos tienen el mismo host.

///

Luego, usando el frontend, puedes hacer que el agente de IA haga cosas en tu nombre.

# CoreOS stuff
/cluster/libvirt-coreos/coreos_*.img

# Downloaded Kubernetes binary release
/kubernetes/

# direnv .envrc files
.envrc

# Local agent override file
AGENTS.override.md
# TODO: remove once there is a merged AGENTS.md
AGENTS.md

# Downloaded kubernetes binary release tar ball
kubernetes.tar.gz

# Phony test files used as part of coverage generation
zz_generated_*_test.go

import jetbrains.buildServer.configs.kotlin.buildSteps.script

object DownloadGitRepoToEc2Agent : BuildType({
    val id = "Util_DownloadGitRepoToEc2Agent"
    name = "Download Git Repo to EC2 Agent"
    description = "Do nothing but downloading gradle/gradle repo to EC2 agents"

    applyDefaultSettings(artifactRuleOverride = "")

    params {
        param("defaultBranchName", "master")
    }

    steps {
        script {

Imagine que você desenvolve uma forma de executar um agente de IA local.

Ele fornece uma API em

```
http://localhost:8000/v1/agents/multivac
```

Há também um frontend em

```
http://localhost:8000
```

/// tip | Dica

Observe que ambos têm o mesmo host.

///

Usando o frontend, você pode fazer o agente de IA executar ações em seu nome.

## Example Attack { #example-attack }

Imagine you build a way to run a local AI agent.

It provides an API at

```
http://localhost:8000/v1/agents/multivac
```

There's also a frontend at

```
http://localhost:8000
```

/// tip

Note that both have the same host.

///

Then using the frontend you can make the AI agent do things on your behalf.

Bir API sunuyor:

```
http://localhost:8000/v1/agents/multivac
```

Ayrıca bir frontend var:

```
http://localhost:8000
```

/// tip | İpucu

İkisinin de host’u aynıdır.

///

Frontend’i kullanarak AI agent’a sizin adınıza işler yaptırabiliyorsunuz.

}

fun Requirements.requiresNotEc2Agent() {
    doesNotContain("teamcity.agent.name", "ec2")
    // US region agents have name "EC2-XXX"
    doesNotContain("teamcity.agent.name", "EC2")
}

/**
 * We have some "shared" host where a Linux build agent and a Windows build agent
 * both run on the same bare metal. Some builds require exclusive access to the

## Exemple d’attaque { #example-attack }

Imaginez que vous mettiez au point un moyen d’exécuter un agent IA local.

Il expose une API à l’adresse

```
http://localhost:8000/v1/agents/multivac
```

Il y a aussi un frontend à l’adresse

```
http://localhost:8000
```

/// tip | Astuce

Notez qu’ils ont le même hôte.

///

- 且沒有送出任何身分驗證憑證

這種攻擊主要與以下情境相關：

- 應用在本機（例如 `localhost`）或內部網路中執行
- 並且應用沒有任何身分驗證，假設同一個網路中的任何請求都可被信任

## 攻擊範例 { #example-attack }

假設你打造了一個在本機執行 AI 代理（AI agent）的方法。

它提供一個 API：

```
http://localhost:8000/v1/agents/multivac
```

同時也有一個前端：

```
http://localhost:8000
```

/// tip | 提示

請注意兩者的主機（host）相同。

///

接著你可以透過前端讓 AI 代理代你執行動作。