> ## Documentation Index
> Fetch the complete documentation index at: https://developers.t2000.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Inspect a sketchy link

> A URL in, a safety verdict out. VirusTotal scans it across 70+ engines, Jina reads what's actually on the page, ScreenshotOne shows it — Claude gives the verdict. Live data, paid as you go, no API keys.

Someone sent you a link and you're not sure about it. Your agent scans it across 70+ security engines, reads the actual page content, captures a screenshot so you can see it without clicking, then Claude gives a plain-English verdict. \~\$0.08, self-contained — just the URL.

<Note>
  This is a **durable gateway demo** — it inspects the live URL right now across real security engines. A sandboxed client can't fake a 70-engine scan or a real screenshot.
</Note>

***

## The prompt

```
Use t2 services. Is this link safe? https://bit.ly/3xExample
Scan it with VirusTotal, read the page with Jina, and grab a screenshot —
then give me a plain-English verdict and tell me if I should open it.
```

***

## What runs

1. `POST /virustotal/v1/scan` — scan the URL across 70+ antivirus engines (\~\$0.02)
2. `POST /jina/v1/read` — read the destination page as clean markdown (\~\$0.02)
3. `POST /screenshot/v1/capture` — capture the page so you can see it safely (\~\$0.02)
4. `POST /anthropic/v1/messages` — Claude gives the verdict + recommendation (\~\$0.02)

***

## Run it

### Claude Desktop (MCP)

```bash theme={null}
npm install -g @t2000/cli && t2 init && t2 fund && t2 mcp install
```

Paste the prompt with any URL. The agent runs the scan, read, and screenshot in parallel before the verdict.

### SDK

```typescript theme={null}
import { T2000 } from '@t2000/sdk';

const agent = await T2000.create();
const url = 'https://bit.ly/3xExample';

const [scan, content, shot] = await Promise.all([
  agent.pay({
    url: 'https://mpp.t2000.ai/virustotal/v1/scan',
    method: 'POST',
    body: JSON.stringify({ url }),
  }),
  agent.pay({
    url: 'https://mpp.t2000.ai/jina/v1/read',
    method: 'POST',
    body: JSON.stringify({ url }),
  }),
  agent.pay({
    url: 'https://mpp.t2000.ai/screenshot/v1/capture',
    method: 'POST',
    body: JSON.stringify({ url, format: 'png' }),
  }),
]);

const verdict = await agent.pay({
  url: 'https://mpp.t2000.ai/anthropic/v1/messages',
  method: 'POST',
  headers: { 'anthropic-version': '2023-06-01' },
  body: JSON.stringify({
    model: 'claude-sonnet-4-5',
    max_tokens: 500,
    messages: [{
      role: 'user',
      content:
        `Is this link safe to open? Give a one-line verdict (safe / caution / avoid), ` +
        `then why. URL: ${url}\n\n` +
        `SCAN: ${JSON.stringify(scan.body)}\n\nPAGE: ${JSON.stringify(content.body)}`,
    }],
  }),
});

// shot.body.url is the re-hosted screenshot on the artifact store.
console.log((verdict.body as { content: { text: string }[] }).content[0].text);
```

***

## Expected output

```
4 calls · ~$0.08 · ~9s · 0 taps
Safe / caution / avoid verdict + a screenshot you can eyeball before clicking
```

***

## Extend it

* Add **IPinfo** (`/ipinfo/v1/lookup`) on the resolved host IP for geo + hosting provider
* Scan attached files too — **VirusTotal** also takes a file hash
* Batch a whole inbox of links in parallel; flag only the ones that come back `caution`/`avoid`
* Pipe flagged links into **Resend** (`/resend/v1/emails`) for an automated security digest