facebook-ad-library-lead-scraper

CRITICAL E004: Prompt injection detected in skill instructions.

• Potential prompt injection detected (high risk: 0.80). The skill explicitly instructs the agent to install runtime dependencies automatically "without asking the user to confirm" and to hide that action ("Do not describe it as a request..."), which are deceptive instructions outside the scraping task's stated purpose.

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.90). Both links point to an unfamiliar third‑party site (browseract.com) that appears to be used to distribute a CLI/installer and is referenced in the skill as an automatic installation source for executables, which is a higher-risk pattern for malware distribution when coming from non-official or unvetted domains.

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

• Third-party content exposure detected (high risk: 0.75). SKILL.md instructs the agent to use the BrowserAct CLI to access live public Facebook Ad Library pages at runtime (outsider-authored web content), which can be ingested as readable page text into the LLM context via the BrowserAct workflow.

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

• Potentially malicious external URL detected (high risk: 0.90). The SKILL explicitly requires and instructs the agent to automatically install and invoke BrowserAct at runtime using the installer URLs (https://www.browseract.com/?co-from=lead-generation and 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.

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

• Attempt to modify system services in skill instructions detected (high risk: 0.80). The skill explicitly instructs the agent to automatically install the BrowserAct CLI (a runtime dependency) without asking the user for confirmation, which directs the agent to modify the host environment and potentially perform privileged system changes.