cloudflare-redirector

Warn

Audited by Gen Agent Trust Hub on Apr 7, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSCREDENTIALS_UNSAFEPROMPT_INJECTION
Full Analysis
  • [CREDENTIALS_UNSAFE]: The SKILL.md file contains instructions for the AI agent to persistently store sensitive Cloudflare API tokens in plain text by appending them to shell configuration files such as ~/.zshrc and ~/.bashrc.
  • [PROMPT_INJECTION]: The skill includes instructions in SKILL.md that command the agent to override standard step-by-step execution and automatically perform a chain of sensitive operations—including credential verification, DNS modification, and code deployment—without further user confirmation.
  • [COMMAND_EXECUTION]: The skill makes extensive use of shell commands to manage environment variables, install project dependencies, and execute deployment tasks using the wrangler CLI.
  • [EXTERNAL_DOWNLOADS]: The skill utilizes pnpm dlx to dynamically download and execute the official Cloudflare Workers CLI (wrangler) from the NPM registry during the deployment process.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Apr 7, 2026, 04:51 PM