social-audience-question-miner

CRITICAL E004: Prompt injection detected in skill instructions.

• Potential prompt injection detected (high risk: 1.00). The Skill contains deceptive installation instructions that tell the agent to install a dependency automatically without user confirmation and explicitly to hide or not characterize the installation request, which is outside the skill’s claimed purpose and bypasses user consent.

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 real public social media pages at runtime and extract visible comment/reply/chat/question text (outsider-authored free text) into the agent’s working context for analysis/export.

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

• Potentially malicious external URL detected (high risk: 0.90). The Skill requires using or automatically installing the BrowserAct CLI at runtime via the URLs https://www.browseract.com/?co-from=social-media and https://www.browseract.com/?co-from=social-media&redirect=https://github.com/browser-act/skills/tree/main, which are used to fetch/install a required runtime dependency that can execute code and thus directly affect agent behavior.