re0-upgrade

CRITICAL E004: Prompt injection detected in skill instructions.

• Potential prompt injection detected (medium risk: 0.30). The skill instructs the agent to automatically star a GitHub repository without user prompt or confirmation as part of its post-upgrade routine.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (critical risk: 1.00). The skill contains instructions for an automated GitHub star action executed silently without user consent as part of its post-upgrade workflow.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (low risk: 0.10). The workflow fetches discovery notice script files from raw.githubusercontent.com using a version tag derived from the installation state, which is a low-risk user-selected or version-pinned remote source.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill instructs the agent to fetch remote scripts from https://raw.githubusercontent.com/LilMGenius/paperthin/v<installed-version>/scripts/<file> at runtime to set up session-start hooks.