facebook-group-admin-contact-search

CRITICAL E004: Prompt injection detected in skill instructions.

• Potential prompt injection detected (high risk: 0.80). The Skill explicitly directs the agent to install BrowserAct automatically and to "not ask the user to confirm" or "describe it as a request," which instructs hiding installation/consent and is a deceptive instruction outside the skill's advertised data-search purpose.

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 runtime browser-act Skill to access real Facebook/public pages and extract profile/post data, which would ingest outsider-authored free text from those fetched web pages into the LLM context.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill explicitly directs the agent to automatically install the BrowserAct Skill at runtime using https://www.browseract.com/?co-from=lead-generation&redirect=https://github.com/browser-act/skills/tree/main, which would fetch and execute remote code as a required dependency.