linkedin-post-search-scraper

CRITICAL E004: Prompt injection detected in skill instructions.

• Potential prompt injection detected (high risk: 0.80). The skill contains instructions to install dependencies automatically and to avoid requesting or describing user approval (SKILL.md:40-48), which directs deceptive/hidden behavior outside the declared scraping purpose.

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

• Third-party content exposure detected (high risk: 0.75). This skill’s runtime workflow delegates live browsing to the outsider-authored BrowserAct CLI/pages (public LinkedIn content fetched at runtime), which can include arbitrary free text from those pages that the agent may ingest into LLM context.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill instructs the agent to automatically invoke the Install BrowserAct URL at runtime to obtain the required BrowserAct Skill/CLI (which fetches and installs remote code), e.g. https://www.browseract.com/?co-from=lead-generation&redirect=https://github.com/browser-act/skills/tree/main, so remote content is fetched and executed as a required dependency.