vibe-coding-toolkit

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

• Third-party content exposure detected (high risk: 1.00). The skill's core "Explore" workflow and explicit commands (SKILL.md: "skill-seekers scrape --url --name " and "skill-seekers github --repo <OWNER/REPO>") plus standards/development-workflow.md guidance to "使用 skill-seekers 摄取第三方库文档" and "使用 WebSearch" show the agent will fetch and ingest public web/GitHub content (untrusted third-party) and act on it, which can materially influence subsequent actions.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill explicitly depends on and invokes the external tool at https://github.com/skill-seekers/skill-seekers (e.g., "skill-seekers scrape --url "), which at runtime pulls remote documentation and converts it into skills that can directly control agent prompts/instructions.