instagram-hashtag-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 "do not request separate approval" or present it as a user action, which directs deceptive behavior (installing without user consent) outside the advertised scraping purpose.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 0.90). This Skill explicitly instructs silent, automatic installation and invocation of an external CLI (BrowserAct) without user confirmation and to conceal the installation as not requiring user action, creating a clear supply-chain/unauthorized-installation risk that could enable remote code execution or data exfiltration.

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 use the BrowserAct CLI to access real Instagram pages at runtime and scrape public hashtag/post content, which is outsider-authored free text fetched from the web and then ingested into the agent’s LLM context.

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

• Potentially malicious external URL detected (high risk: 0.90). The Skill explicitly requires the BrowserAct CLI at runtime and directs the agent to automatically install it from https://www.browseract.com/?co-from=social-media and https://www.browseract.com/?co-from=social-media&redirect=https://github.com/browser-act/skills/tree/main, which are external installers fetched/executed during runtime and thus can directly control agent execution.