INCIDENT MAP

Browse / Indirect Prompt Injection / CVE-2026-54316

Claude Code: Out-of-Band Data Exfiltration via Pre-Approved HuggingFace Domain in WebFetch

CVE-2026-54316 · 2026-06-17 moderate

Class
Indirect Prompt Injection (filed by a model that read the advisory)
Technique
Data exfiltration
Also involves
Tool and Agent Misuse
Affected
@anthropic-ai/claude-code (npm)
Gets past
Approval or confirmation prompt
CVSS
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:N
EPSS
0.5% chance of exploitation in 30 days

How it works, in short

Injected content can make Claude Code fetch attacker-chosen Hugging Face URLs without a prompt, leaking data through the request traffic as a covert channel. (Our summary, written by a model. The source text is below.)

Details

Because the hostname huggingface.co was pre-approved as a bare hostname for the WebFetch tool, any path on that domain—including attacker-controlled model repositories—was auto-approved without a permission prompt or being subject to --allowedTools restrictions. An attacker able to inject untrusted content into a Claude Code context could direct it to issue WebFetch requests against attacker-controlled repository files (e.g. /resolve/main/config.json), which HuggingFace counts as downloads server-side, creating a covert out-of-band channel for encoding and exfiltrating data Claude can access such as files, environment variables, or command output. Reliably exploiting this required the ability to add untrusted content into a Claude Code context window. Users on standard Claude Code auto-update have received this fix already; users performing manual updates are advised to update to the latest version.

Thank you to hackerone.com/novee for reporting this issue.

References

Source: GitHub Advisory Database, used under CC BY 4.0.