Claude Code: Out-of-Band Data Exfiltration via Pre-Approved HuggingFace Domain in WebFetch
Key facts
- Attack vector
- Network
- Privileges required
- None
- User interaction
- Passive
- Attack complexity
- Low
- Attack requirements
- Present
- Impact type
- Information disclosure
- Confidentiality · integrity · availability
- High · None · None
- Fix status
- Partial Red Hat: 1 without a fix
Attack profile from CVSS 4.0 · CNA. Fix status from CNA, GitHub and Red Hat. Impact type from CWE-200.
Summary
Claude Code is an agentic coding tool. From 0.2.54 until 2.1.163, 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. This vulnerability is fixed in 2.1.163.
Why it matters now
SSVC decision points
CISA ADP · SSVC 2.0.3
- Exploitation
- None
- Automatable
- No
- Technical impact
- Partial
The decision (Track, Attend, Act) also depends on your mission and well-being impact; see CISA's SSVC guide.
Affected products
| Vendor | Product | Affected versions | Fixed in | Status | Source |
|---|---|---|---|---|---|
| anthropics | claude-code | 0.2.54 to before 2.1.163 | 2.1.163 | Patch available | CNA |
| — | @anthropic-ai/claude-code (npm) | 0.2.54 to before 2.1.163 | 2.1.163 | Patch available | GHSA-fg94-h982-f3mm |
- claude-code Vendoranthropics Affected 0.2.54 to before 2.1.163 Fixed in 2.1.163 Patch available CNA
- @anthropic-ai/claude-code (npm) Vendor— Affected 0.2.54 to before 2.1.163 Fixed in 2.1.163 Patch available GHSA-fg94-h982-f3mm
CPE configurations NVD · 1
| CPE | Versions | Role | Vendor |
|---|---|---|---|
| cpe:2.3:a:anthropic:claude_code:*:*:*:*:*:node.js:*:* | 0.2.54 to before 2.1.163 | Vulnerable | anthropics |
- cpe:2.3:a:anthropic:claude_code:*:*:*:*:*:node.js:*:* Versions 0.2.54 to before 2.1.163 Vulnerable ·anthropics
Remediation
- Fix · CNA claude-code (anthropics) Fixed in 2.1.163
- Fix · GHSA-fg94-h982-f3mm @anthropic-ai/claude-code (npm) Fixed in 2.1.163 Advisory →
Vendor assessments
Red Hat
Red Hat rating: Moderate. Red Hat's page for this CVE →
| Product | Status | Plan or fix |
|---|---|---|
| Red Hat OpenShift Dev Spaces | Affected | No fix yet: Fix deferred |
Timeline
- 2026-10-02 Threatist record created from CVE-2026-54316
- 2026-10-03 Alias added: GHSA-fg94-h982-f3mm
References
More anthropics Claude Code vulnerabilities
- TI-2025-40558 Claude Code Command Validation Bypass Allows Arbitrary Code Execution CVSS 4.0HIGH 8.7
- TI-2025-36683 Claude Code vulnerable to arbitrary code execution caused by maliciously configured git email CVSS 4.0HIGH 8.7
- TI-2025-37066 Claude Code's startup trust dialog could lead to Command Execution attack CVSS 4.0HIGH 8.7
- TI-2026-17554 Claude Code Leaks Data via Malicious Environment Configuration Before Trust Confirmation CVSS 4.0MEDIUM 5.3
- TI-2025-40093 @anthropic-ai/claude-code has Sed Command Validation Bypass that Allows Arbitrary File Writes CVSS 4.0HIGH 8.7
- TI-2025-36422 Claude Code rg command had Command Injection that allowed bypass of user approval prompt for command execution CVSS 4.0HIGH 8.7
Sources
- CVE-2026-54316 · CVE Program
- CVE-2026-54316 · NVD
- FIRST EPSS
- GHSA-fg94-h982-f3mm · GitHub Advisory Database · CC BY 4.0
- CVE-2026-54316 · Red Hat security data · CC BY 4.0
This product uses the NVD API but is not endorsed or certified by the NVD. Sources and licences