vercel-deploy

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). The package contains an explicit hardcoded external deploy endpoint that uploads the entire project tarball and instructions to escalate network permissions—indicative of intentional data exfiltration risk and potential backdoor access to user projects.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The deployment script posts the project tarball at runtime to https://codex-deploy-skills.vercel.sh/api/deploy, which triggers remote deployment execution and the returned JSON (preview/claim URLs) directly controls the agent's workflow.

MEDIUM W013: Attempt to modify system services in skill instructions.

• Attempt to modify system services in skill instructions detected (high risk: 0.80). The skill explicitly directs the agent to request and use "escalated" sandbox permissions / network access to rerun the deploy command (bypassing sandboxing/security restrictions), which is an instruction to obtain elevated privileges and thus can compromise the host state.