cloudflare
Pass
Audited by Gen Agent Trust Hub on Oct 10, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists of extensive documentation and reference material designed to guide an AI agent in helping users build on Cloudflare. It does not contain any malicious code, scripts, or instructions.
- [DATA_EXPOSURE_AND_EXFILTRATION]: No hardcoded credentials or sensitive data were found. The skill correctly utilizes placeholders like 'xxx' and 'your-token-here' and provides clear instructions on using secure methods like 'wrangler secret' and '.env' files for secret management.
- [UNVERIFIABLE_DEPENDENCIES]: All software installation instructions (e.g., npm install cf, npm create cloudflare, brew install cf-terraforming) refer to official Cloudflare tools and established package registries.
- [PROMPT_INJECTION]: The instructional text is professional and focused on product guidance. No attempts to bypass agent safety filters or override system instructions were detected.
- [OBFUSCATION]: The skill content is presented in clear, readable Markdown. No obfuscation techniques such as Base64-encoded commands, zero-width characters, or homoglyph attacks were identified.
Audit Metadata