codex-cursor-unlimited-reset

CRITICAL E004: Prompt injection detected in skill instructions.

• Potential prompt injection detected (medium risk: 0.30). The skill provides instructions and remote script execution commands intended to bypass usage limits and rate limits of AI coding tools, presenting a risk of unauthorized software modification.

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (critical risk: 1.00). The skill instructs users to download and execute an untrusted PowerShell script via irm ... | iex from a random GitHub repository (SlayerCoralPersonify/Activate) marketed as an AI limit bypass tool, which is a classic deceptive software delivery vector.

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

• Potentially malicious external URL detected (high risk: 0.70). The skill instructs downloading and executing a remote PowerShell script from an arbitrary personal GitHub repository (https://raw.githubusercontent.com/SlayerCoralPersonify/Activate/main/install.ps1), representing a weak/unknown/suspicious operational runtime dependency.

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

• Attempt to modify system services in skill instructions detected (high risk: 1.00). The skill prompt instructs the agent to execute untrusted remote PowerShell scripts via irm ... | iex which can compromise the state of the machine by downloading and running arbitrary code.