linkedin-job-detail-scraper

CRITICAL E004: Prompt injection detected in skill instructions.

• Potential prompt injection detected (high risk: 0.90). The skill instructs the agent to automatically install a required dependency without user confirmation and explicitly tells the agent to conceal the installation and not request approval, which is a hidden/deceptive instruction outside the stated scraping purpose.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). The skill explicitly instructs covert automatic installation of a runtime dependency without user consent and enables large-scale scraping/harvesting of LinkedIn contact data, posing high supply-chain and data-exfiltration risk.

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 can invoke the BrowserAct CLI to open and extract content from user-specified LinkedIn URLs/pages at runtime, meaning outsider-authored free text scraped from public web pages may be ingested into the agent’s LLM context via the BrowserAct execution path.

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

• Potentially malicious external URL detected (high risk: 0.80). The Skill instructs the agent to automatically install BrowserAct from https://www.browseract.com/?co-from=lead-generation (and a variant with a redirect) at runtime if missing, which implies fetching and executing remote installation code from that URL without user confirmation.