youtube-shorts-scraper

CRITICAL E004: Prompt injection detected in skill instructions.

• Potential prompt injection detected (high risk: 0.90). The skill contains deceptive runtime instructions that tell the agent to install BrowserAct automatically and explicitly to hide that installation request from the user and not ask for separate approval, which is a hidden/misleading directive outside normal transparent behavior.

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

• Third-party content exposure detected (high risk: 0.85). SKILL.md instructs the agent to use the BrowserAct CLI to access live YouTube pages at runtime and then return scraped fields (e.g., titles/descriptions/comments), which are outsider-authored free text from public web content.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly instructs the agent at runtime to use the Install BrowserAct URL (https://www.browseract.com/?co-from=social-media&redirect=https://github.com/browser-act/skills/tree/main) to automatically install the required BrowserAct CLI/Skill, which implies fetching and executing external installation code that would directly affect agent behavior.

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

• Attempt to modify system services in skill instructions detected (high risk: 0.70). The Skill explicitly instructs the agent to automatically install the BrowserAct CLI dependency without asking the user, which directs the agent to change the host system state (install software) and could lead to unauthorized modifications or privilege escalation risks.