g2-company-lead-scraper

CRITICAL E004: Prompt injection detected in skill instructions.

• Potential prompt injection detected (high risk: 0.80). The Skill instructs the agent to install runtime dependencies automatically and explicitly tells it to hide/avoid asking the user for confirmation, which is a deceptive instruction outside the advertised lead-scraping behavior.

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

• Third-party content exposure detected (high risk: 0.85). This skill’s required workflow invokes the browser-act Skill to access and interact with live public web pages at runtime, so outsider-authored free text from those fetched pages can be ingested into the agent’s LLM context via the BrowserAct browsing/extraction path.

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 automatically install the BrowserAct CLI/Skill at runtime using the Install BrowserAct link (https://www.browseract.com/?co-from=lead-generation&redirect=https://github.com/browser-act/skills/tree/main), which would fetch and execute remote installation code and thus is a runtime dependency that can run remote code.