cloudflare-deploy

Warn

Audited by Socket on Sep 14, 2026

2 alerts found:

Anomalyx2
AnomalyLOW
references/sandbox/api.md

This fragment provides documentation for a powerful sandbox SDK that supports arbitrary command/code execution, filesystem modification, external bucket mounting with write capability, and network exposure including WebSocket proxying. There is no direct evidence of embedded malware or obfuscation in the snippet itself. However, the combination of execution + networking + storage write + output streaming makes the overall system high-impact and dual-use; security largely depends on strict authorization, input validation (especially around request-derived routing), egress controls, and secret-handling/redaction in the surrounding implementation.

Confidence: 65%Severity: 60%
AnomalyLOW
references/pages-functions/patterns.md

No definitive on-code backdoor (no eval/Function/dynamic execution, no process spawning, no credential theft logic) is shown in this fragment. The highest concern is the unconditional hardcoded outbound POST to an external webhook executed via ctx.waitUntil on incoming requests, which is consistent with tracking or potential data exfiltration. Secondary security weaknesses include wildcard CORS, returning err.message to clients, and writing uploaded content to object storage using an unsanitized, client-controlled filename. Treat the webhook behavior as the primary supply-chain risk and verify what payload is sent, whether it is conditional, and whether sensitive data can reach it; also add validation/sanitization for uploads, tighten CORS, and avoid error-message disclosure.

Confidence: 58%Severity: 64%
Audit Metadata
Analyzed At
Sep 14, 2026, 09:36 PM
Package URL
pkg:socket/skills-sh/openai%2Fskills%2Fcloudflare-deploy%2F@3eec7c659b0ac25384d21eb6361163224a0baa8bf440cf2656d799bab2c507e5
Security Audit — socket — cloudflare-deploy