claude-code-tutor

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

• Third-party content exposure detected (high risk: 0.90). This skill explicitly instructs the agent to fetch and scrape public third-party web content (e.g., "use agent-browser to scrape the pricing table from competitor.com", "/clone-website example.com", and using Chrome MCP to control a browser to visit external sites) and to install skills from GitHub (npx skills add...), so untrusted, user-generated public content is read and used to drive actions.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly requires installing a third-party skill at runtime (e.g., "npx skills add arjunkmrm/recall"), which fetches remote content that defines/controls agent behavior (SKILL.md) and is used as a required dependency for module C1, so this is a runtime external dependency that can control prompts/instructions.