convert-to-markdown

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 1.00). This is a raw shell script hosted in a personal GitHub repo by an unknown user and the skill recommends piping it to sh (curl | sh), which is high-risk because such remote scripts can contain arbitrary malicious commands.

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

• Third-party content exposure detected (high risk: 0.80). SKILL.md explicitly instructs the agent to fetch and extract content from arbitrary URLs using the ezycopy CLI (Phase 2) and to present that extracted markdown to the user, so untrusted public/web user-generated content is ingested and can influence agent behavior (e.g., extraction retries with --browser in Phase 1/3).

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

• Potentially malicious external URL detected (high risk: 1.00). The skill includes a runtime install instruction that fetches and pipes a shell script from https://raw.githubusercontent.com/gupsammy/EzyCopy/main/install.sh (curl ... | sh), which would execute remote code and is relied on to provide the required ezycopy CLI.