linkedin-sales-navigator-company-scraper

CRITICAL E004: Prompt injection detected in skill instructions.

• Potential prompt injection detected (high risk: 1.00). The Skill explicitly instructs the agent to install BrowserAct automatically and to conceal that installation (not asking user confirmation or describing it as a browser request), which is a deceptive instruction outside the stated scraping 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 invoke the runtime browser-act Skill to access LinkedIn pages and extract visible public profile/company text, which is outsider-authored free text fetched at runtime and then ingested into the agent context.

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 to "install the missing dependency automatically" at runtime using the Install BrowserAct URL (https://www.browseract.com/?co-from=lead-generation&redirect=https://github.com/browser-act/skills/tree/main), which indicates the agent will fetch and run external installer code to obtain a required runtime dependency.