Claude Gets VirusTotal Threat Intelligence Capabilities for AI-Assisted Malware and URL Analysis

Blog WriterCybersecurity News - Original News Source is cybersecuritynews.com


Claude users can connect VirusTotal threat intelligence to their investigations, bringing file, URL, domain, and IP reports into an AI workflow. The connection lets analysts review detection findings, analysis dates, coverage, and source links without treating an AI answer as a security verdict.

The integration uses Model Context Protocol, or MCP, to connect Claude with VirusTotal AI, known as VTAI. Users authenticate through browser OAuth without providing a personal VirusTotal API key. Organization rules may require an owner to approve the connector before employees can use it.

For security teams, the main benefit is evidence review. An analyst can provide a file hash or suspicious address and ask Claude to explain the report. This brings the checks described in Cyber Security News’ VirusTotal research guide into a conversation, while keeping the underlying findings available for review.

Report lookups remain separate from submissions. Only an explicit user request authorizes file uploads, URL submissions or domain and IP reanalysis. Before uploading a file, the service checks its verified hash. Original bytes are uploaded only after confirmed absence, not after a permission error or failed lookup.

The approach extends a workflow covered in the VirusTotal ChatGPT integration. However, missing reports and zero detections do not establish safety. Analysts still need to consider report age and coverage rather than reading an empty result as approval to open a file or link.

Local File Analysis and Privacy Limits

According to VirusTotal’s Claude Code documentation, an optional plugin adds a threat intelligence skill and local file hook. The hook checks permitted paths, original bytes, size, and SHA-256 before transfer. It blocks common credential filenames for human review. Claude chat does not run this local hook, and the remote connector cannot read local paths.

When explicitly requested, submit_attachment can retrieve original file bytes from a public HTTPS download link. This does not establish native Claude attachment support. VirusTotal documents authenticated testing in Claude Code, while noting that hosted Claude chat report workflows, renewal, and writes remain unverified.

The official connection guide lists free access at 60 queries per minute and 1,000 per UTC day, shared across account connections. Repeated reports and explicit analysis reads count each time. New file contributions and owned receipt recovery use no query quota, but separate contribution limits apply.

Standard submissions are shared with VirusTotal’s community and security partners, so private documents, internal code, credentials and personal data require permission before disclosure. URL queries expose the complete address, including parameters and fragments.

Users can revoke access through Your connections, but removing the plugin does not erase prior submissions or receipts already held by the service.

Give your SOC the alert context to stops incidents quickly. Integrate Threat Intel from 16K SOCs.

The post Claude Gets VirusTotal Threat Intelligence Capabilities for AI-Assisted Malware and URL Analysis appeared first on Cyber Security News.