Certificat AgentReady.md vérifié
Émis le sig: 496415cea2c01595 Vérifier →

URL analysée

https://ai.virustotal.com

Mesuré la semaine dernière

Analyser une autre URL

Score AI-Ready

Excellent

sur 100

Économie de tokens

Tokens HTML 5590
Tokens Markdown 1018
Économie 82%

Détail du score

Accessibilité 88/100
Visibilité IA 88/100
Données structurées 93/100
HTML sémantique 96/100
Efficacité du contenu 94/100

Protocoles émergents

2 sur 6 détectés

Endpoints well-known recherchés par les agents IA. Détecté signifie qu’un agent peut découvrir et se connecter automatiquement à votre service.

  • OAuth Protected Resource RFC 9728
    /.well-known/oauth-protected-resource
  • OAuth Discovery RFC 8414
    /.well-known/oauth-authorization-server
  • MCP Server Card SEP-1649 draft
    /.well-known/mcp/server-card.json
  • A2A Agent Card A2A v1.0
    /.well-known/agent-card.json
  • API Catalog RFC 9727
    /.well-known/api-catalog
    • 2 API(s)
  • Agent Skills index Discovery RFC v0.2.0 draft
    /.well-known/agent-skills/index.json
    • 1 skill(s)
    • vtai-core

Cela ne reste pas en place. La surveillance n'existe pas encore : inscrivez-vous, on vous préviendra.

Vous êtes sur la liste ! Nous vous préviendrons lors du lancement.

Ce que nous avons mesuré 2/3 OG tags present

Balises Open Graph manquantes ou incomplètes. Les balises OG aident les AI agents (et les plateformes sociales) à comprendre le titre, la description et l’image de votre page.

Comment implémenter

Ajoutez les balises meta og:title, og:description et og:image dans le <head> de votre page.

Collez-le dans un agent de code pour appliquer le correctif
Tokens Markdown: 1018
Threat intelligence, ready for your agent

Check files, links and infrastructure. Submit files for analysis. Give your agent evidence to act on.

Free VTAI access within quotas. No VirusTotal API key needed.

## Know what you are about to trust.

Evidence for downloads, investigations and autonomous workflows.

01### Check an indicator

Read the existing report for a file hash, URL, domain or IP. A hash lookup does not upload the file.

`get_file_report` · `get_url_report` · `get_domain_report` · `get_ip_report`

02### Analyze a file

Submit the file and keep its SHA-256 and receipt. Stdio can read a local file; HTTP accepts inline content.

`submit_file` · `submit_local_file (stdio only)`

03### Resume an analysis

Recover the same account's receipt without resending. Read its analysis ID when available; an uncertain outcome can remain unknown.

`get_submission` · `get_analysis`

**Capabilities, sharing and limits**

Seven common MCP tools; stdio adds local file submission. Inline submissions accept up to 24,000,000 decoded bytes; local files up to 32,000,000 bytes. Submissions use standard public sharing without a per-call confirmation. The agent applies your task's authorization and data policy. URL, domain and IP lookups read existing reports.

A missing report, a pending analysis or zero detections is not a guarantee of safety. Keep analysis dates and coverage with the evidence. [Read the tool guide](https://ai.virustotal.com/?format=markdown).

## Protection inside your runtime.

Prefer an integrated plugin? Start here.

VT Sentinel · IDE

### Google Antigravity extension

Check supported IDE downloads with VirusTotal. Start with a protected terminal and inspect coverage with the extension self-test.

`antigravity --install-extension virustotal.vt-sentinel`

[Open VSX & setup](https://open-vsx.org/extension/virustotal/vt-sentinel)

**Activate and verify**

Open a workspace. Run VT Sentinel: Run Self-Test, then VT Sentinel: Open Protected Terminal from the Command Palette. This uses the Antigravity IDE CLI, not agy.

Run antigravity --list-extensions --show-versions and inspect the self-test. Follow its trust or restart instructions before relying on interception.

HTTP(S) interception only. Remote workspaces cover new IDE terminals and tasks. Some systems need a trust approval or managed restart; the self-test shows the effective scope.

VT Sentinel · OpenClaw

### OpenClaw plugin

Add file reputation tools and automatic artifact checks. Inspect and adjust scanning, upload and blocking policies from the agent.

`openclaw plugins install clawhub:openclaw-plugin-vt-sentinel`

[Setup & configuration](https://github.com/king-tero/VT-sentinel)

**Activate and verify**

Run openclaw gateway restart, then openclaw plugins list. Restart only the gateway you intend to update.

Call vt\_sentinel\_status to inspect active policy, then vt\_check\_hash with a known file hash for a first reputation lookup. Status alone does not verify access to VTAI.

Protection follows the configured scanning, upload and blocking policies. Instruction files default to hash-only; sensitive-file upload behavior is configurable.

VirusTotal · Hermes

### Hermes plugin

Give Hermes file and hash reputation tools, plus compact advisory context about the artifacts it observes.

`hermes plugins install king-tero/hermes-virustotal --enable`

[Setup & configuration](https://github.com/king-tero/hermes-virustotal)

**Activate and verify**

Run hermes plugins list and start a new Hermes session. If using a messaging gateway, restart it with hermes gateway restart. The --enable flag activates the plugin without an interactive enable prompt.

Call vt\_check\_hash with a known file hash, or vt\_check\_file with a path in the active execution environment. The file check hashes the file without uploading it.

Advisory by default; exact-malicious enforcement is optional. Binary-content upload behavior has separate configuration. Text is not auto-uploaded and archives are opt-in.

[Installation, activation and first use ↗](https://ai.virustotal.com/install.md)

Direct integration

## Just need the API?

VTAI turns VirusTotal intelligence into compact responses for agent workflows. Register once, reuse your token and call the API from your own tools.

`GET /api/v3/domains/virustotal.com Authorization: Bearer <VTAI token>`

Use a protected credential file or secret store. The guide includes executable requests for reports, submissions and recovery.

[OpenAPI schema](https://ai.virustotal.com/openapi.json) · [Integration catalog](https://ai.virustotal.com/?format=json)

## A starting point for every agent.

Readable without JavaScript.
VirusTotal AI | Threat intelligence for your agent               [Skip to content](https://ai.virustotal.com/#main)

[VirusTotal AI](https://ai.virustotal.com/)

Threat intelligence, ready for your agent

# VirusTotal intelligence for your AI agent.

Check files, links and infrastructure. Submit files for analysis. Give your agent evidence to act on.

Free VTAI access within quotas. No VirusTotal API key needed.

[Connect your agent](https://ai.virustotal.com/#connect)Copy setup prompt[Read agent instructions](https://ai.virustotal.com/install.md)

### Agy

Antigravity CLI

Local stdio[Connect Agy ↗](https://ai.virustotal.com/connect/mcp?client=agy)

### Claude Code

Your existing Claude login

HTTP or stdio[Connect Claude Code ↗](https://ai.virustotal.com/connect/mcp?client=claude)

### Codex

Your existing Codex login

HTTP or stdio[Connect Codex ↗](https://ai.virustotal.com/connect/mcp?client=codex)

## Know what you are about to trust.

Evidence for downloads, investigations and autonomous workflows.

01### Check an indicator

Read the existing report for a file hash, URL, domain or IP. A hash lookup does not upload the file.

`get_file_report` · `get_url_report` · `get_domain_report` · `get_ip_report`

02### Analyze a file

Submit the file and keep its SHA-256 and receipt. Stdio can read a local file; HTTP accepts inline content.

`submit_file` · `submit_local_file (stdio only)`

03### Resume an analysis

Recover the same account's receipt without resending. Read its analysis ID when available; an uncertain outcome can remain unknown.

`get_submission` · `get_analysis`

**Capabilities, sharing and limits**

Seven common MCP tools; stdio adds local file submission. Inline submissions accept up to 24,000,000 decoded bytes; local files up to 32,000,000 bytes. Submissions use standard public sharing without a per-call confirmation. The agent applies your task's authorization and data policy. URL, domain and IP lookups read existing reports.

A missing report, a pending analysis or zero detections is not a guarantee of safety. Keep analysis dates and coverage with the evidence. [Read the tool guide](https://ai.virustotal.com/?format=markdown).

## Protection inside your runtime.

Prefer an integrated plugin? Start here.

VT Sentinel · IDE

### Google Antigravity extension

Check supported IDE downloads with VirusTotal. Start with a protected terminal and inspect coverage with the extension self-test.

`antigravity --install-extension virustotal.vt-sentinel`Copy

[Open VSX & setup](https://open-vsx.org/extension/virustotal/vt-sentinel)

**Activate and verify**

Open a workspace. Run VT Sentinel: Run Self-Test, then VT Sentinel: Open Protected Terminal from the Command Palette. This uses the Antigravity IDE CLI, not agy.

Run antigravity --list-extensions --show-versions and inspect the self-test. Follow its trust or restart instructions before relying on interception.

HTTP(S) interception only. Remote workspaces cover new IDE terminals and tasks. Some systems need a trust approval or managed restart; the self-test shows the effective scope.

VT Sentinel · OpenClaw

### OpenClaw plugin

Add file reputation tools and automatic artifact checks. Inspect and adjust scanning, upload and blocking policies from the agent.

`openclaw plugins install clawhub:openclaw-plugin-vt-sentinel`Copy

[Setup & configuration](https://github.com/king-tero/VT-sentinel)

**Activate and verify**

Run openclaw gateway restart, then openclaw plugins list. Restart only the gateway you intend to update.

Call vt\_sentinel\_status to inspect active policy, then vt\_check\_hash with a known file hash for a first reputation lookup. Status alone does not verify access to VTAI.

Protection follows the configured scanning, upload and blocking policies. Instruction files default to hash-only; sensitive-file upload behavior is configurable.

VirusTotal · Hermes

### Hermes plugin

Give Hermes file and hash reputation tools, plus compact advisory context about the artifacts it observes.

`hermes plugins install king-tero/hermes-virustotal --enable`Copy

[Setup & configuration](https://github.com/king-tero/hermes-virustotal)

**Activate and verify**

Run hermes plugins list and start a new Hermes session. If using a messaging gateway, restart it with hermes gateway restart. The --enable flag activates the plugin without an interactive enable prompt.

Call vt\_check\_hash with a known file hash, or vt\_check\_file with a path in the active execution environment. The file check hashes the file without uploading it.

Advisory by default; exact-malicious enforcement is optional. Binary-content upload behavior has separate configuration. Text is not auto-uploaded and archives are opt-in.

[Installation, activation and first use ↗](https://ai.virustotal.com/install.md)

Direct integration

## Just need the API?

VTAI turns VirusTotal intelligence into compact responses for agent workflows. Register once, reuse your token and call the API from your own tools.

[Read the API quickstart](https://ai.virustotal.com/skills/BASIC.md)[API reference ↗](https://ai.virustotal.com/docs)

`GET /api/v3/domains/virustotal.com Authorization: Bearer <VTAI token>`

Use a protected credential file or secret store. The guide includes executable requests for reports, submissions and recovery.

[OpenAPI schema](https://ai.virustotal.com/openapi.json) · [Integration catalog](https://ai.virustotal.com/?format=json)

## A starting point for every agent.

Readable without JavaScript.

[**Installation guide**Choose an integration and follow its setup instructions.`/install.md`](https://ai.virustotal.com/install.md) [**Agent overview**Decide when and how to use VirusTotal.`/llms.txt`](https://ai.virustotal.com/llms.txt) [**VTAI API guide**Register, authenticate, query, submit and recover with REST.`/skills/BASIC.md`](https://ai.virustotal.com/skills/BASIC.md) [**OpenAPI**Exact request and response schemas.`/openapi.json`](https://ai.virustotal.com/openapi.json)

Téléversez ce fichier sous /index.md sur votre serveur pour que les AI agents puissent accéder à une version propre de votre page. Vous pouvez également configurer la négociation de contenu Accept: text/markdown pour le servir automatiquement.

Notre recommandation

Télécharger llms.txt
# ai.virustotal.com

> Connect Agy, Claude Code or Codex to VirusTotal. Check files, URLs, domains and IPs, submit files for analysis, or integrate with VTAI directly. Free access within quotas.

## Documentation
- [For agents](https://ai.virustotal.com/install.md)
- [Read the API quickstart](https://ai.virustotal.com/skills/BASIC.md)
- [API reference ↗](https://ai.virustotal.com/docs)
- [Agent overviewDecide when and how to use VirusTotal./llms.txt](https://ai.virustotal.com/llms.txt)

## Main
- [VirusTotal intelligence for your AI agent.](https://ai.virustotal.com): Connect Agy, Claude Code or Codex to VirusTotal. Check files, URLs, domains and IPs, submit files for analysis, or inte…
- [VirusTotal AI](https://ai.virustotal.com/)
- [For agents](https://ai.virustotal.com/install.md)
- [Sitemap](https://ai.virustotal.com/sitemap.xml)

Le llms.txt complet nécessite une analyse de tout le domaine (bientôt disponible)

Téléversez ce fichier vers https://ai.virustotal.com/llms.txt à la racine de votre domaine. Les AI agents comme ChatGPT, Claude et Perplexity consultent ce fichier pour comprendre la structure de votre site.

Ce site possède déjà un fichier llms.txt.

Format valide
# VirusTotal AI

> Threat intelligence for agents making decisions about files, links and infrastructure.

Use VirusTotal to check an unfamiliar download, investigate a suspicious URL/domain/IP, analyze a file, or resume a pending analysis. Reports return compact evidence; a hash lookup does not upload a file. Access is free within VTAI quotas and requires a VTAI token. No existing VirusTotal API key is needed.

## Install

- [Choose an integration](/install.md): MCP, runtime plugin or direct API.
- [Agy setup](/connect/mcp?client=agy&format=markdown): Local stdio. [Local files / stdio](/connect/mcp?client=agy&transport=stdio&format=markdown).
- [Claude Code setup](/connect/mcp?client=claude&format=markdown): HTTP or stdio. [Local files / stdio](/connect/mcp?client=claude&transport=stdio&format=markdown).
- [Codex setup](/connect/mcp?client=codex&format=markdown): HTTP or stdio. [Local files / stdio](/connect/mcp?client=codex&transport=stdio&format=markdown).

## Use

- [MCP tools and workflows](/?format=markdown): look up, submit, recover.
- [VTAI API skill](/skills/BASIC.md): executable REST examples and result interpretation.
- [OpenAPI](/openapi.json): exact schemas.
- [Integration catalog](/?format=json): structured capabilities and setup links.

## Runtime plugins

- [Google Antigravity extension](https://open-vsx.org/extension/virustotal/vt-sentinel): Check supported IDE downloads with VirusTotal. Start with a protected terminal and inspect coverage with the extension self-test.
- [OpenClaw plugin](https://github.com/king-tero/VT-sentinel): Add file reputation tools and automatic artifact checks. Inspect and adjust scanning, upload and blocking policies from the agent.
- [Hermes plugin](https://github.com/king-tero/hermes-virustotal): Give Hermes file and hash reputation tools, plus compact advisory context about the artifacts it observes.

## Operation

Reuse your VTAI credential; use Bearer or x-apikey, never both. Files use standard public sharing when submitted. Recover an uncertain submission by SHA-256 with get_submission before reading its analysis ID with get_analysis; do not automatically resubmit. Unknown can remain unknown. URL/domain/IP lookups read existing reports. Unknown, pending, errors and zero detections are not guarantees of safety.

Accessibilité

Contenu disponible sans JavaScript (100/100)

Content available without JavaScript

Le contenu apparaît tôt dans le HTML (50/100)

Main content starts at 47% of HTML

Taille de page raisonnable (100/100)

Page size: 20KB

Visibilité IA

robots.txt autorise les bots IA (100/100)

All major AI search bots allowed

Support Markdown for Agents (60/100) Application
✓ Accept: text/markdown ✗ .md URL ✓ <link> tag ✓ Link header
Possède un sitemap.xml (100/100)

Sitemap found

Possède un fichier robots.txt (100/100)

robots.txt exists

Possède un fichier llms.txt (100/100)

llms.txt exists and is valid

A Content-Signal (robots.txt ou en-têtes HTTP) (60/100)
✓ robots.txt ✗ HTTP header ✗ Policy

Données structurées

Possède Schema.org / JSON-LD (100/100)

JSON-LD found: WebPage

Possède des balises Open Graph (67/100)

2/3 OG tags present

Possède une meta description (100/100)

Meta description: 171 chars

Possède une URL canonique (100/100)

Canonical URL present

Possède un attribut lang (100/100)

lang="en"

HTML sémantique

Hiérarchie de titres correcte (85/100)

1 heading level skip(s)

Utilise les éléments article ou main (100/100)

Has both <article> and <main>

Utilise des éléments HTML sémantiques (100/100)

17 semantic elements, 20 divs (ratio: 46%)

Textes alternatifs descriptifs pour les images (100/100)

No images found

Faible profondeur d’imbrication des divs (100/100)

Avg div depth: 0.3, max: 2

Efficacité du contenu

Bon ratio de réduction de tokens (100/100)

82% token reduction (HTML→Markdown)

Bon ratio contenu/bruit (80/100)

Content ratio: 21.1% (4257 content chars / 20205 HTML bytes)

Poids de page raisonnable (100/100)

HTML size: 20KB

Styles en ligne minimaux (100/100)

0/194 elements with inline styles (0.0%)

{
  "url": "https://ai.virustotal.com",
  "timestamp": 1789092241877,
  "fetch": {
    "mode": "simple",
    "timeMs": 132,
    "htmlSizeBytes": 20205,
    "supportsMarkdown": true,
    "markdownAgents": {
      "contentNegotiation": true,
      "mdUrl": {
        "found": false,
        "url": null
      },
      "linkTag": {
        "found": true,
        "url": "/?format=markdown"
      },
      "linkHeader": {
        "found": true,
        "url": "/?format=markdown"
      },
      "responseHeaders": {
        "contentSignal": null,
        "xMarkdownTokens": null,
        "vary": "Accept"
      },
      "frontmatter": {
        "present": false,
        "fields": [],
        "level": "none"
      },
      "level": "application",
      "contentNegotiationMediaType": "text/markdown",
      "properMediaType": true
    },
    "statusCode": 200
  },
  "extraction": {
    "title": "VirusTotal intelligence for your AI agent.",
    "excerpt": "Connect Agy, Claude Code or Codex to VirusTotal. Check files, URLs, domains and IPs, submit files for analysis, or integrate with VTAI directly. Free access within quotas.",
    "byline": null,
    "siteName": null,
    "lang": "en",
    "contentLength": 4257,
    "metadata": {
      "description": "Connect Agy, Claude Code or Codex to VirusTotal. Check files, URLs, domains and IPs, submit files for analysis, or integrate with VTAI directly. Free access within quotas.",
      "ogTitle": "VirusTotal intelligence for your AI agent.",
      "ogDescription": "Connect Agy, Claude Code or Codex to VirusTotal. Check files, URLs, domains and IPs, submit files for analysis, or integrate with VTAI directly. Free access within quotas.",
      "ogImage": null,
      "ogType": "website",
      "canonical": "https://ai.virustotal.com/",
      "lang": "en",
      "schemas": [
        {
          "@context": "https://schema.org",
          "@type": "WebPage",
          "name": "VirusTotal AI",
          "url": "https://ai.virustotal.com/",
          "description": "Connect Agy, Claude Code or Codex to VirusTotal. Check files, URLs, domains and IPs, submit files for analysis, or integrate with VTAI directly. Free access within quotas.",
          "about": {
            "@type": "SoftwareApplication",
            "name": "VirusTotal MCP",
            "applicationCategory": "SecurityApplication",
            "url": "https://ai.virustotal.com/connect/mcp",
            "downloadUrl": "https://ai.virustotal.com/install.md",
            "description": "Read VirusTotal reports, submit files for analysis and recover the results through MCP. Free access uses a VTAI token and shared quotas; no VirusTotal API key is needed."
          }
        }
      ],
      "robotsMeta": null,
      "author": null,
      "generator": null,
      "markdownAlternateHref": "/?format=markdown"
    }
  },
  "markdown": "Threat intelligence, ready for your agent\n\nCheck files, links and infrastructure. Submit files for analysis. Give your agent evidence to act on.\n\nFree VTAI access within quotas. No VirusTotal API key needed.\n\n## Know what you are about to trust.\n\nEvidence for downloads, investigations and autonomous workflows.\n\n01### Check an indicator\n\nRead the existing report for a file hash, URL, domain or IP. A hash lookup does not upload the file.\n\n`get_file_report` · `get_url_report` · `get_domain_report` · `get_ip_report`\n\n02### Analyze a file\n\nSubmit the file and keep its SHA-256 and receipt. Stdio can read a local file; HTTP accepts inline content.\n\n`submit_file` · `submit_local_file (stdio only)`\n\n03### Resume an analysis\n\nRecover the same account's receipt without resending. Read its analysis ID when available; an uncertain outcome can remain unknown.\n\n`get_submission` · `get_analysis`\n\n**Capabilities, sharing and limits**\n\nSeven common MCP tools; stdio adds local file submission. Inline submissions accept up to 24,000,000 decoded bytes; local files up to 32,000,000 bytes. Submissions use standard public sharing without a per-call confirmation. The agent applies your task's authorization and data policy. URL, domain and IP lookups read existing reports.\n\nA missing report, a pending analysis or zero detections is not a guarantee of safety. Keep analysis dates and coverage with the evidence. [Read the tool guide](https://ai.virustotal.com/?format=markdown).\n\n## Protection inside your runtime.\n\nPrefer an integrated plugin? Start here.\n\nVT Sentinel · IDE\n\n### Google Antigravity extension\n\nCheck supported IDE downloads with VirusTotal. Start with a protected terminal and inspect coverage with the extension self-test.\n\n`antigravity --install-extension virustotal.vt-sentinel`\n\n[Open VSX & setup](https://open-vsx.org/extension/virustotal/vt-sentinel)\n\n**Activate and verify**\n\nOpen a workspace. Run VT Sentinel: Run Self-Test, then VT Sentinel: Open Protected Terminal from the Command Palette. This uses the Antigravity IDE CLI, not agy.\n\nRun antigravity --list-extensions --show-versions and inspect the self-test. Follow its trust or restart instructions before relying on interception.\n\nHTTP(S) interception only. Remote workspaces cover new IDE terminals and tasks. Some systems need a trust approval or managed restart; the self-test shows the effective scope.\n\nVT Sentinel · OpenClaw\n\n### OpenClaw plugin\n\nAdd file reputation tools and automatic artifact checks. Inspect and adjust scanning, upload and blocking policies from the agent.\n\n`openclaw plugins install clawhub:openclaw-plugin-vt-sentinel`\n\n[Setup & configuration](https://github.com/king-tero/VT-sentinel)\n\n**Activate and verify**\n\nRun openclaw gateway restart, then openclaw plugins list. Restart only the gateway you intend to update.\n\nCall vt\\_sentinel\\_status to inspect active policy, then vt\\_check\\_hash with a known file hash for a first reputation lookup. Status alone does not verify access to VTAI.\n\nProtection follows the configured scanning, upload and blocking policies. Instruction files default to hash-only; sensitive-file upload behavior is configurable.\n\nVirusTotal · Hermes\n\n### Hermes plugin\n\nGive Hermes file and hash reputation tools, plus compact advisory context about the artifacts it observes.\n\n`hermes plugins install king-tero/hermes-virustotal --enable`\n\n[Setup & configuration](https://github.com/king-tero/hermes-virustotal)\n\n**Activate and verify**\n\nRun hermes plugins list and start a new Hermes session. If using a messaging gateway, restart it with hermes gateway restart. The --enable flag activates the plugin without an interactive enable prompt.\n\nCall vt\\_check\\_hash with a known file hash, or vt\\_check\\_file with a path in the active execution environment. The file check hashes the file without uploading it.\n\nAdvisory by default; exact-malicious enforcement is optional. Binary-content upload behavior has separate configuration. Text is not auto-uploaded and archives are opt-in.\n\n[Installation, activation and first use ↗](https://ai.virustotal.com/install.md)\n\nDirect integration\n\n## Just need the API?\n\nVTAI turns VirusTotal intelligence into compact responses for agent workflows. Register once, reuse your token and call the API from your own tools.\n\n`GET /api/v3/domains/virustotal.com Authorization: Bearer <VTAI token>`\n\nUse a protected credential file or secret store. The guide includes executable requests for reports, submissions and recovery.\n\n[OpenAPI schema](https://ai.virustotal.com/openapi.json) · [Integration catalog](https://ai.virustotal.com/?format=json)\n\n## A starting point for every agent.\n\nReadable without JavaScript.\n",
  "fullPageMarkdown": "VirusTotal AI | Threat intelligence for your agent               [Skip to content](https://ai.virustotal.com/#main)\n\n[VirusTotal AI](https://ai.virustotal.com/)\n\nThreat intelligence, ready for your agent\n\n# VirusTotal intelligence for your AI agent.\n\nCheck files, links and infrastructure. Submit files for analysis. Give your agent evidence to act on.\n\nFree VTAI access within quotas. No VirusTotal API key needed.\n\n[Connect your agent](https://ai.virustotal.com/#connect)Copy setup prompt[Read agent instructions](https://ai.virustotal.com/install.md)\n\n### Agy\n\nAntigravity CLI\n\nLocal stdio[Connect Agy ↗](https://ai.virustotal.com/connect/mcp?client=agy)\n\n### Claude Code\n\nYour existing Claude login\n\nHTTP or stdio[Connect Claude Code ↗](https://ai.virustotal.com/connect/mcp?client=claude)\n\n### Codex\n\nYour existing Codex login\n\nHTTP or stdio[Connect Codex ↗](https://ai.virustotal.com/connect/mcp?client=codex)\n\n## Know what you are about to trust.\n\nEvidence for downloads, investigations and autonomous workflows.\n\n01### Check an indicator\n\nRead the existing report for a file hash, URL, domain or IP. A hash lookup does not upload the file.\n\n`get_file_report` · `get_url_report` · `get_domain_report` · `get_ip_report`\n\n02### Analyze a file\n\nSubmit the file and keep its SHA-256 and receipt. Stdio can read a local file; HTTP accepts inline content.\n\n`submit_file` · `submit_local_file (stdio only)`\n\n03### Resume an analysis\n\nRecover the same account's receipt without resending. Read its analysis ID when available; an uncertain outcome can remain unknown.\n\n`get_submission` · `get_analysis`\n\n**Capabilities, sharing and limits**\n\nSeven common MCP tools; stdio adds local file submission. Inline submissions accept up to 24,000,000 decoded bytes; local files up to 32,000,000 bytes. Submissions use standard public sharing without a per-call confirmation. The agent applies your task's authorization and data policy. URL, domain and IP lookups read existing reports.\n\nA missing report, a pending analysis or zero detections is not a guarantee of safety. Keep analysis dates and coverage with the evidence. [Read the tool guide](https://ai.virustotal.com/?format=markdown).\n\n## Protection inside your runtime.\n\nPrefer an integrated plugin? Start here.\n\nVT Sentinel · IDE\n\n### Google Antigravity extension\n\nCheck supported IDE downloads with VirusTotal. Start with a protected terminal and inspect coverage with the extension self-test.\n\n`antigravity --install-extension virustotal.vt-sentinel`Copy\n\n[Open VSX & setup](https://open-vsx.org/extension/virustotal/vt-sentinel)\n\n**Activate and verify**\n\nOpen a workspace. Run VT Sentinel: Run Self-Test, then VT Sentinel: Open Protected Terminal from the Command Palette. This uses the Antigravity IDE CLI, not agy.\n\nRun antigravity --list-extensions --show-versions and inspect the self-test. Follow its trust or restart instructions before relying on interception.\n\nHTTP(S) interception only. Remote workspaces cover new IDE terminals and tasks. Some systems need a trust approval or managed restart; the self-test shows the effective scope.\n\nVT Sentinel · OpenClaw\n\n### OpenClaw plugin\n\nAdd file reputation tools and automatic artifact checks. Inspect and adjust scanning, upload and blocking policies from the agent.\n\n`openclaw plugins install clawhub:openclaw-plugin-vt-sentinel`Copy\n\n[Setup & configuration](https://github.com/king-tero/VT-sentinel)\n\n**Activate and verify**\n\nRun openclaw gateway restart, then openclaw plugins list. Restart only the gateway you intend to update.\n\nCall vt\\_sentinel\\_status to inspect active policy, then vt\\_check\\_hash with a known file hash for a first reputation lookup. Status alone does not verify access to VTAI.\n\nProtection follows the configured scanning, upload and blocking policies. Instruction files default to hash-only; sensitive-file upload behavior is configurable.\n\nVirusTotal · Hermes\n\n### Hermes plugin\n\nGive Hermes file and hash reputation tools, plus compact advisory context about the artifacts it observes.\n\n`hermes plugins install king-tero/hermes-virustotal --enable`Copy\n\n[Setup & configuration](https://github.com/king-tero/hermes-virustotal)\n\n**Activate and verify**\n\nRun hermes plugins list and start a new Hermes session. If using a messaging gateway, restart it with hermes gateway restart. The --enable flag activates the plugin without an interactive enable prompt.\n\nCall vt\\_check\\_hash with a known file hash, or vt\\_check\\_file with a path in the active execution environment. The file check hashes the file without uploading it.\n\nAdvisory by default; exact-malicious enforcement is optional. Binary-content upload behavior has separate configuration. Text is not auto-uploaded and archives are opt-in.\n\n[Installation, activation and first use ↗](https://ai.virustotal.com/install.md)\n\nDirect integration\n\n## Just need the API?\n\nVTAI turns VirusTotal intelligence into compact responses for agent workflows. Register once, reuse your token and call the API from your own tools.\n\n[Read the API quickstart](https://ai.virustotal.com/skills/BASIC.md)[API reference ↗](https://ai.virustotal.com/docs)\n\n`GET /api/v3/domains/virustotal.com Authorization: Bearer <VTAI token>`\n\nUse a protected credential file or secret store. The guide includes executable requests for reports, submissions and recovery.\n\n[OpenAPI schema](https://ai.virustotal.com/openapi.json) · [Integration catalog](https://ai.virustotal.com/?format=json)\n\n## A starting point for every agent.\n\nReadable without JavaScript.\n\n[**Installation guide**Choose an integration and follow its setup instructions.`/install.md`](https://ai.virustotal.com/install.md) [**Agent overview**Decide when and how to use VirusTotal.`/llms.txt`](https://ai.virustotal.com/llms.txt) [**VTAI API guide**Register, authenticate, query, submit and recover with REST.`/skills/BASIC.md`](https://ai.virustotal.com/skills/BASIC.md) [**OpenAPI**Exact request and response schemas.`/openapi.json`](https://ai.virustotal.com/openapi.json)\n",
  "markdownStats": {
    "images": 0,
    "links": 7,
    "tables": 0,
    "codeBlocks": 0,
    "headings": 10
  },
  "tokens": {
    "htmlTokens": 5590,
    "markdownTokens": 1018,
    "reduction": 4572,
    "reductionPercent": 82
  },
  "score": {
    "score": 91,
    "grade": "A",
    "rubricVersion": 3,
    "dimensions": {
      "accessibility": {
        "score": 88,
        "weight": 30,
        "grade": "B",
        "checks": {
          "content_without_js": {
            "score": 100,
            "weight": 55,
            "evidence": "proven",
            "details": "Content available without JavaScript"
          },
          "fast_content_position": {
            "score": 50,
            "weight": 25,
            "evidence": "plausible",
            "details": "Main content starts at 47% of HTML"
          },
          "reasonable_page_size": {
            "score": 100,
            "weight": 20,
            "evidence": "plausible",
            "details": "Page size: 20KB"
          }
        }
      },
      "aiDiscoverability": {
        "score": 88,
        "weight": 25,
        "grade": "B",
        "checks": {
          "robots_allows_ai_bots": {
            "score": 100,
            "weight": 35,
            "evidence": "proven",
            "details": "All major AI search bots allowed"
          },
          "supports_markdown_negotiation": {
            "score": 60,
            "weight": 20,
            "evidence": "plausible",
            "details": "Application level — Content negotiation, <link> tag, Link header",
            "mechanisms": {
              "negotiation": true,
              "mdUrl": false,
              "linkTag": true,
              "linkHeader": true
            }
          },
          "has_sitemap": {
            "score": 100,
            "weight": 15,
            "evidence": "plausible",
            "details": "Sitemap found"
          },
          "has_robots_txt": {
            "score": 100,
            "weight": 10,
            "evidence": "plausible",
            "details": "robots.txt exists"
          },
          "has_llms_txt": {
            "score": 100,
            "weight": 10,
            "evidence": "speculative",
            "details": "llms.txt exists and is valid"
          },
          "has_content_signals": {
            "score": 60,
            "weight": 10,
            "evidence": "speculative",
            "details": "robots.txt: ai-input=yes, ai-train=no, search=yes",
            "mechanisms": {
              "robotsTxt": true,
              "httpHeader": false,
              "policy": false
            }
          }
        }
      },
      "structuredData": {
        "score": 93,
        "weight": 20,
        "grade": "A",
        "checks": {
          "has_schema_org": {
            "score": 100,
            "weight": 35,
            "evidence": "proven",
            "details": "JSON-LD found: WebPage"
          },
          "has_open_graph": {
            "score": 67,
            "weight": 20,
            "evidence": "plausible",
            "details": "2/3 OG tags present"
          },
          "has_meta_description": {
            "score": 100,
            "weight": 20,
            "evidence": "plausible",
            "details": "Meta description: 171 chars"
          },
          "has_canonical_url": {
            "score": 100,
            "weight": 15,
            "evidence": "plausible",
            "details": "Canonical URL present"
          },
          "has_lang_attribute": {
            "score": 100,
            "weight": 10,
            "evidence": "plausible",
            "details": "lang=\"en\""
          }
        }
      },
      "semanticHtml": {
        "score": 96,
        "weight": 15,
        "grade": "A",
        "checks": {
          "proper_heading_hierarchy": {
            "score": 85,
            "weight": 30,
            "evidence": "plausible",
            "details": "1 heading level skip(s)"
          },
          "uses_article_or_main": {
            "score": 100,
            "weight": 25,
            "evidence": "plausible",
            "details": "Has both <article> and <main>"
          },
          "semantic_elements": {
            "score": 100,
            "weight": 20,
            "evidence": "plausible",
            "details": "17 semantic elements, 20 divs (ratio: 46%)"
          },
          "meaningful_alt_texts": {
            "score": 100,
            "weight": 15,
            "evidence": "plausible",
            "details": "No images found"
          },
          "low_div_nesting": {
            "score": 100,
            "weight": 10,
            "evidence": "speculative",
            "details": "Avg div depth: 0.3, max: 2"
          }
        }
      },
      "contentEfficiency": {
        "score": 94,
        "weight": 10,
        "grade": "A",
        "checks": {
          "token_reduction_ratio": {
            "score": 100,
            "weight": 40,
            "evidence": "speculative",
            "details": "82% token reduction (HTML→Markdown)"
          },
          "content_to_noise_ratio": {
            "score": 80,
            "weight": 30,
            "evidence": "speculative",
            "details": "Content ratio: 21.1% (4257 content chars / 20205 HTML bytes)"
          },
          "reasonable_page_weight": {
            "score": 100,
            "weight": 20,
            "evidence": "speculative",
            "details": "HTML size: 20KB"
          },
          "minimal_inline_styles": {
            "score": 100,
            "weight": 10,
            "evidence": "speculative",
            "details": "0/194 elements with inline styles (0.0%)"
          }
        }
      }
    }
  },
  "recommendations": [
    {
      "id": "add_open_graph",
      "priority": "medium",
      "category": "structuredData",
      "titleKey": "rec.add_open_graph.title",
      "descriptionKey": "rec.add_open_graph.description",
      "howToKey": "rec.add_open_graph.howto",
      "howToStepKeys": null,
      "effort": "quick-win",
      "estimatedImpact": 1.3,
      "maxImpact": 4,
      "evidence": "plausible",
      "checkScore": 67,
      "checkDetails": "2/3 OG tags present"
    }
  ],
  "llmsTxtPreview": "# ai.virustotal.com\n\n> Connect Agy, Claude Code or Codex to VirusTotal. Check files, URLs, domains and IPs, submit files for analysis, or integrate with VTAI directly. Free access within quotas.\n\n## Documentation\n- [For agents](https://ai.virustotal.com/install.md)\n- [Read the API quickstart](https://ai.virustotal.com/skills/BASIC.md)\n- [API reference ↗](https://ai.virustotal.com/docs)\n- [Agent overviewDecide when and how to use VirusTotal./llms.txt](https://ai.virustotal.com/llms.txt)\n\n## Main\n- [VirusTotal intelligence for your AI agent.](https://ai.virustotal.com): Connect Agy, Claude Code or Codex to VirusTotal. Check files, URLs, domains and IPs, submit files for analysis, or inte…\n- [VirusTotal AI](https://ai.virustotal.com/)\n- [For agents](https://ai.virustotal.com/install.md)\n- [Sitemap](https://ai.virustotal.com/sitemap.xml)\n\n",
  "llmsTxtExisting": "# VirusTotal AI\n\n> Threat intelligence for agents making decisions about files, links and infrastructure.\n\nUse VirusTotal to check an unfamiliar download, investigate a suspicious URL/domain/IP, analyze a file, or resume a pending analysis. Reports return compact evidence; a hash lookup does not upload a file. Access is free within VTAI quotas and requires a VTAI token. No existing VirusTotal API key is needed.\n\n## Install\n\n- [Choose an integration](/install.md): MCP, runtime plugin or direct API.\n- [Agy setup](/connect/mcp?client=agy&format=markdown): Local stdio. [Local files / stdio](/connect/mcp?client=agy&transport=stdio&format=markdown).\n- [Claude Code setup](/connect/mcp?client=claude&format=markdown): HTTP or stdio. [Local files / stdio](/connect/mcp?client=claude&transport=stdio&format=markdown).\n- [Codex setup](/connect/mcp?client=codex&format=markdown): HTTP or stdio. [Local files / stdio](/connect/mcp?client=codex&transport=stdio&format=markdown).\n\n## Use\n\n- [MCP tools and workflows](/?format=markdown): look up, submit, recover.\n- [VTAI API skill](/skills/BASIC.md): executable REST examples and result interpretation.\n- [OpenAPI](/openapi.json): exact schemas.\n- [Integration catalog](/?format=json): structured capabilities and setup links.\n\n## Runtime plugins\n\n- [Google Antigravity extension](https://open-vsx.org/extension/virustotal/vt-sentinel): Check supported IDE downloads with VirusTotal. Start with a protected terminal and inspect coverage with the extension self-test.\n- [OpenClaw plugin](https://github.com/king-tero/VT-sentinel): Add file reputation tools and automatic artifact checks. Inspect and adjust scanning, upload and blocking policies from the agent.\n- [Hermes plugin](https://github.com/king-tero/hermes-virustotal): Give Hermes file and hash reputation tools, plus compact advisory context about the artifacts it observes.\n\n## Operation\n\nReuse your VTAI credential; use Bearer or x-apikey, never both. Files use standard public sharing when submitted. Recover an uncertain submission by SHA-256 with get_submission before reading its analysis ID with get_analysis; do not automatically resubmit. Unknown can remain unknown. URL/domain/IP lookups read existing reports. Unknown, pending, errors and zero detections are not guarantees of safety.",
  "emergingProtocols": {
    "oauthProtectedResource": {
      "exists": false,
      "url": "https://ai.virustotal.com/.well-known/oauth-protected-resource"
    },
    "oauthDiscovery": {
      "exists": false,
      "url": "https://ai.virustotal.com/.well-known/oauth-authorization-server"
    },
    "mcpServerCard": {
      "exists": false,
      "url": "https://ai.virustotal.com/.well-known/mcp/server-card.json",
      "draft": true
    },
    "a2aAgentCard": {
      "exists": false,
      "url": "https://ai.virustotal.com/.well-known/agent-card.json"
    },
    "apiCatalog": {
      "exists": true,
      "url": "https://ai.virustotal.com/.well-known/api-catalog",
      "contentType": "application/linkset+json; profile=\"https://www.rfc-editor.org/info/rfc9727\"",
      "validMediaType": true,
      "apis": 2
    },
    "agentSkills": {
      "exists": true,
      "url": "https://ai.virustotal.com/.well-known/agent-skills/index.json",
      "draft": true,
      "schema": "https://schemas.agentskills.io/discovery/0.2.0/schema.json",
      "skills": 1,
      "names": [
        "vtai-core"
      ]
    },
    "count": 2,
    "total": 6
  },
  "botAccess": {
    "probed": true,
    "bot": "OAI-SearchBot",
    "controlStatus": 200,
    "botStatus": 200,
    "discriminates": false,
    "refusedAsBot": false,
    "edge": null,
    "verifiable": false,
    "detail": "This origin answers OAI-SearchBot exactly as it answers any other client (200). No edge-level filtering of AI crawlers observed."
  },
  "snippets": [
    {
      "id": "add_open_graph",
      "title": "Add missing Open Graph tags",
      "description": "Open Graph tags control how your page looks when shared on social media and how AI platforms preview your URL in answers.",
      "language": "html",
      "code": "<meta property=\"og:image\" content=\"https://yoursite.com/og-image.jpg\">\n<meta property=\"og:url\" content=\"https://ai.virustotal.com\">\n<meta property=\"og:type\" content=\"website\">",
      "filename": "<head>",
      "stacks": [
        {
          "id": "html",
          "label": "HTML <head>",
          "language": "html",
          "filename": "<head>",
          "code": "<meta property=\"og:image\" content=\"https://yoursite.com/og-image.jpg\">\n<meta property=\"og:url\" content=\"https://ai.virustotal.com\">\n<meta property=\"og:type\" content=\"website\">"
        },
        {
          "id": "wordpress",
          "label": "WordPress",
          "language": "php",
          "filename": "functions.php",
          "code": "<?php\n// Quick Open Graph tags without a plugin (skip if Yoast / Rank Math is active)\nadd_action('wp_head', function () {\n    if (!is_singular()) return;\n    $post = get_queried_object();\n    $title = get_the_title($post);\n    $desc  = get_the_excerpt($post) ?: wp_trim_words(strip_tags($post->post_content), 30);\n    $image = get_the_post_thumbnail_url($post, 'large') ?: 'https://yoursite.com/og-image.jpg';\n    $url   = get_permalink($post);\n    printf('<meta property=\"og:title\" content=\"%s\">' . \"\\n\", esc_attr($title));\n    printf('<meta property=\"og:description\" content=\"%s\">' . \"\\n\", esc_attr($desc));\n    printf('<meta property=\"og:image\" content=\"%s\">' . \"\\n\", esc_url($image));\n    printf('<meta property=\"og:url\" content=\"%s\">' . \"\\n\", esc_url($url));\n    echo '<meta property=\"og:type\" content=\"article\">' . \"\\n\";\n}, 5);"
        },
        {
          "id": "nextjs",
          "label": "Next.js",
          "language": "typescript",
          "filename": "app/page.tsx",
          "code": "// Next.js App Router — Metadata API\nimport type { Metadata } from 'next';\n\nexport const metadata: Metadata = {\n  title: \"VirusTotal intelligence for your AI agent.\",\n  description: \"Connect Agy, Claude Code or Codex to VirusTotal. Check files, URLs, domains and IPs, submit files for analysis, or integrate with VTAI directly. Free access within quotas.\",\n  openGraph: {\n    title: \"VirusTotal intelligence for your AI agent.\",\n    description: \"Connect Agy, Claude Code or Codex to VirusTotal. Check files, URLs, domains and IPs, submit files for analysis, or integrate with VTAI directly. Free access within quotas.\",\n    url: \"https://ai.virustotal.com\",\n    images: [\"https://yoursite.com/og-image.jpg\"],\n    type: 'website',\n  },\n};"
        }
      ]
    }
  ]
}

Utilisez notre API pour obtenir ceci de manière programmatique (bientôt disponible)

Ce JSON est à usage interne — contrairement au Markdown et au llms.txt, il n’est pas destiné à être téléversé sur votre site. Conservez-le comme référence pour suivre l’évolution de votre score, partagez-le avec votre équipe de développement ou intégrez-le dans votre pipeline CI/CD.

Partagez vos résultats

Ou demandez à une IA de vous aider à l'améliorer

Envie d'un deuxième avis ?

Cloudflare propose aussi un scanner gratuit, et il pose une question différente. Il note ce que votre site publie pour que les agents l'appellent : MCP server cards, Agent Skills, catalogues d'API, enregistrements DNS. Nous notons si un agent peut atteindre votre contenu, le lire et le comprendre. Un site peut réussir l'un et échouer l'autre : attendez-vous à deux chiffres différents, ce sont les réponses à deux questions différentes, et les deux méritent d'être connues.

Analyser ai.virustotal.com avec Cloudflare

Intégrez votre badge

Ajoutez ce badge à votre site. Il se met à jour automatiquement lorsque votre score de préparation à l’IA change.

AgentReady.md score for ai.virustotal.com
Script Recommandé
<script src="https://agentready.md/badge.js" data-id="cd81d46f-b23a-4a39-b7b0-e7b814a7622e" data-domain="ai.virustotal.com"></script>
Markdown
[![AgentReady.md score for ai.virustotal.com](https://agentready.md/badge/ai.virustotal.com.svg)](https://agentready.md/fr/r/ai.virustotal.com)

Bientôt : Analyse de domaine complet

Explorez l’ensemble de votre domaine, générez un llms.txt et surveillez votre score de préparation IA dans le temps. Rejoignez la liste d’attente pour être informé.

Vous êtes sur la liste ! Nous vous préviendrons lors du lancement.