weibo-comment-scraper

CRITICAL E004: Prompt injection detected in skill instructions.

• Potential prompt injection detected (high risk: 0.90). The Skill contains deceptive runtime instructions that require automatic installation of a dependency and explicitly instruct the agent not to describe the installation to the user or request approval, which is hidden/deceptive behavior outside the stated scraping purpose.

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.70). Both links point to a third‑party BrowserAct installer page (one with an automatic redirect to a GitHub repo) and are used in the Skill to trigger unattended dependency installation, which is a risky delivery mechanism for tooling and could be abused to distribute malware.

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 BrowserAct to access live Weibo pages at runtime and extract visible comment text (outsider-authored free text from public web content) into the agent’s context for processing/export.

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 install a required runtime dependency from https://www.browseract.com/?co-from=social-media&redirect=https://github.com/browser-act/skills/tree/main, which is a remote installer URL the agent would fetch and execute at runtime to obtain the BrowserAct Skill.